[1] The Attorney General of Canada appeals from an order of the Federal Court (2021 FC 656, McVeigh J.) certifying the action brought on behalf of Joe David Nasogaluak (a minor when the action was commenced) as a class proceeding.

[2] The amended statement of claim describes Mr. Nasogaluak as Aboriginal. He lives in Tuktoyaktuk, Northwest Territories. The pleading alleges that in November 2017, when he was 15, he was arrested near Tuktoyaktuk by two officers of the Royal Canadian Mounted Police, who then assaulted him and made derogatory statements about his racial origin. As a result, it is alleged, he sustained lasting physical and psychological harm.

[3] Acting through his mother as his litigation guardian, Mr. Nasogaluak commenced a proposed class proceeding in the Federal Court against the federal Crown, as represented by the Attorney General, on behalf of a putative class described as

[4] The Class Period is defined as the period from January 1, 1928 (when, it is pleaded, the federal Crown first entered into formal Police Services Agreements with the Territories) to the present.

[5] In accordance with Police Services Agreements with each of the Territories, the RCMP operates some 22 detachments in the Northwest Territories, 26 in Nunavut, and 14 in Yukon Territory. It is pleaded that at all material times, the RCMP had exclusive jurisdiction over its Officers in the Territories, who provided, among other things, police services at these detachments and in the community.

[6] The core allegations in the action are set out as follows in paragraphs 4 to 6 of the amended statement of claim:

[7] The relief sought in the amended statement of claim includes, in addition to an order certifying the action as a class proceeding and appointing Mr. Nasogaluak’s litigation guardian as representative plaintiff, declaratory and monetary relief for systemic negligence, breach of fiduciary duty, and breaches of sections 7 and 15 of the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.) 1982, c. 11. There is no free-standing assertion of a cause of action in assault. Rather, paragraph 72 of the amended statement of claim pleads assault (and battery) as one of the categories of injury and damage flowing from the systemic negligence, breach of fiduciary duty, and breaches of the Charter alleged.

[8] As the motion judge described it, the theory of the plaintiff’s case is thus one of “top-down” liability, in which higher-level decisions created the conditions in which lower-level misconduct could occur.

[9] Rule 334.16(1) of the Federal Courts Rules, S.O.R./98-106, sets out five conditions that must be met for a proceeding in the Federal Court to be certified as a class proceeding:

[10] When all five conditions are met, rule 334.16(1) requires that certification be granted: the motion judge “shall, by order, certify the proceeding as a class proceeding […].” The motion judge here determined that the first four conditions were met and that the fifth could be addressed through a conditional order. She granted conditional certification accordingly.

[11] In his appeal to this Court, the Attorney General submits that the motion judge committed reversible errors in her consideration of each of the four conditions for certification that she determined were met. He says that with one qualification—he accepts before this Court that the amended statement of claim discloses a reasonable cause of action for breach of section 15 of the Charter—none of these four conditions are met, and the certification order must therefore be set aside. He also submits, more globally, that the motion judge failed to provide adequate reasons, and failed to carry out the meaningful screening that a motion for certification requires.

[12] With two exceptions, I would reject the grounds of appeal that the Attorney General puts forward. I agree with the Attorney General first, that the pleading of breach of fiduciary duty fails to disclose a reasonable cause of action, and second, that no foundation has been laid for certifying a common question relating to aggregate damages. Therefore, I would allow the appeal in part, set aside the certification order, and remit the certification order to the Federal Court to be amended in light of these two deficiencies.

[13] In explaining why I reach these conclusions, I will start by briefly reviewing the burden resting on the party seeking certification to show that the five conditions are satisfied, and the standard of review on appeal from a certification order. I will next deal in turn with each of the first four conditions for certification, adding further background as necessary for context, and discussing in respect of each condition the decision of the motion judge, the issues raised on appeal, and whether the motion judge erred in determining the condition was met. I do not propose to address separately the Attorney General’s global attack on the rigour of the screening the motion judge carried out and the adequacy of her reasons. These attacks were not pursued separately in oral argument and, in my view, they are best considered as part of the analysis of the more specific errors the Attorney General alleges.

[14] Before I proceed down the path I have laid out, I should note that Mr. Nasogaluak is no longer a minor, and no longer requires a litigation guardian. Therefore, the parties consented following the hearing of the appeal to the removal of his mother as litigation guardian in this Court and to the amendment of the style of cause in this Court to name Mr. Nasogaluak in his own right as respondent. These changes are reflected in the style of cause set out above. The parties had earlier—though also after the hearing of the appeal—consented to a similar order of the motion judge governing the proceeding in the Federal Court. That order also appointed Mr. Nasogaluak as representative plaintiff, thus fulfilling the condition in the certification order. No issue remains as to the fifth condition for certification. I will not address it further.

[15] I should note as well that the material filed in support of the certification motion included affidavits from Mr. Nasogaluak’s mother and from five other members of the putative class. All of the five reside in the Northwest Territories or Nunavut, describe themselves as “visibly Indigenous,” and recount violent interactions with and racist conduct on the part of members of the RCMP. These events are said to have occurred at various locations in the Territories and at various times between 1990 and 2017.

[16] Two of these five deponents were cross-examined, and their accounts challenged. The Attorney General also filed, among other things, the affidavit of one of the two RCMP officers involved in the encounter with Mr. Nasogaluak. His description of what occurred differs markedly from that in Mr. Nasogaluak’s mother’s affidavit.

[17] The record included, in addition, certain expert and other evidence to which I will refer below.

[18] The first certification condition, that the pleadings disclose a reasonable cause of action, is assessed on the same standard that applies on a motion to strike out a pleading. Thus, the question is whether it is plain and obvious, assuming the facts pleaded to be true (unless they are manifestly incapable of being proven), that the pleaded claims have no reasonable prospect of success: Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19 at para. 14; Canada v. Greenwood, 2021 FCA 186 at para. 91, leave to appeal to S.C.C. refused, 39885 (March 17, 2022). A claim that has no reasonable prospect of success will not satisfy the first condition.

[19] No evidence is admissible on this issue. However, the pleading must be read generously, and as it might reasonably be amended to accommodate inadequacies attributable to drafting. Moreover, recognizing that the law is not static, the motion judge must err on the side of permitting a novel but arguable claim to proceed to trial: R. v. Imperial Tobacco Canada Ltd., 2011 SCC 42 at paras. 19-25. This Court has described as “onerous” the burden resting on a defendant seeking to defeat a certification motion on the basis that no reasonable cause of action is pleaded: Greenwood at para. 144.

[20] To satisfy each of the remaining four conditions, a plaintiff seeking certification must adduce evidence that provides “some basis in fact” for concluding that the condition is met: Greenwood at para. 94, citing Hollick v. Toronto (City), 2001 SCC 68, and Pro-Sys Consultants Ltd. v. Microsoft Corporation, 2013 SCC 57, among other authorities. This is a threshold lower than the ordinary civil standard of balance of probabilities. But the motion for certification remains “a meaningful screening device,” and “[t]here must be sufficient facts to satisfy the [motion] judge that the conditions for certification have been met to a degree that should allow the matter to proceed on a class basis without foundering at the merits stage”: Pro-Sys at paras. 101-104.

[21] Whether a pleading discloses a cause of action is a question of law. The standard of appellate review of the motion judge’s decision on the first certification condition is, therefore, correctness: Pioneer Corp. v. Godfrey, 2019 SCC 42 at para. 57; Canada (Attorney General) v. Jost, 2020 FCA 212 at para. 21. On this standard, this Court owes no deference to the court below and is free to substitute the opinion of the motion judge with its own: Housen v. Nikolaisen, 2002 SCC 33 at para. 8.

[22] Absent an extricable question of law, the remaining certification conditions involve questions of fact or of mixed fact and law. The motion judge’s conclusions that these conditions are met are therefore—again, absent an extricable question of law—reviewable on the deferential standard of palpable and overriding error: Jost at para. 21, citing Canada v. John Doe, 2016 FCA 191 at paras. 27-32.

[23] In considering whether the first condition for certification was met, the motion judge correctly referred (at paragraphs 16 and 17 of her reasons) to the “plain and obvious” standard for assessing whether a reasonable cause of action is pleaded. She then dealt sequentially with each of the causes of action asserted in the amended statement of claim: systemic negligence, breach of fiduciary duty, breach of section 7 of the Charter, and breach of section 15. I will do the same.

[24] The relief sought in the amended statement of claim includes, in paragraph 1(b),

[25] Also sought, in paragraph 1(e), is a declaration that Canada is liable to the plaintiff and Class Members for damages caused by, among other things, its negligence in relation to the funding, operation, supervision, control, maintenance, oversight and support of RCMP Detachments and RCMP Officers in the Territories.

[26] There are in addition combined claims for damages for negligence and breach of the Charter in the amount of $500 million, as well as claims for punitive and exemplary damages of $100 million, and claims for interest and costs.

[27] The acts and omissions alleged to have breached the duty of care in negligence—which are also relied upon as breaches of the Crown’s fiduciary duty—are set out in paragraph 60 of the amended statement of claim. They include, among others,

[28] As this Court stated in Greenwood (at para. 153), “the required elements that a plaintiff must establish are the same in all negligence claims, regardless of whether or not they are pursued on a systemic basis.” The Court in Greenwood went on to list these required elements as the Supreme Court set them out in Saadati v. Moorhead, 2017 SCC 28 at para. 13:

[29] Whether a defendant (including a defendant who is a government official) owed a duty of care to the claimant, so that the first element is met, is determined by applying what is known as the Anns/Cooper test: Nelson (City) v. Marchi, 2021 SCC 41 at paras. 16-19. The test has two stages. The question at the first stage is whether the defendant owed the claimant a prima facie duty of care. The answer will be affirmative if there was a relationship of proximity between the claimant and the defendant, meaning that the defendant’s failure to take reasonable care might foreseeably cause loss or harm to the claimant. The question at the second stage is whether there are residual policy concerns outside the parties’ relationship that should negate any prima facie duty of care found to exist at the first stage.

[30] As the Supreme Court clarified in Nelson (at para. 16), “the framework applies differently depending on whether the plaintiff’s claim falls within or is analogous to an established duty of care or whether the claim is novel because proximity has not been recognized before.” In the former situation, it is generally not necessary to proceed to the second stage of the test. However, if the claim is a novel one, “the full two-stage Anns/Cooper framework applies.” In determining whether a claim is analogous to a previously established category of duty, the Court clarified (Nelson at para. 27, internal citations omitted), “a court should be attentive to the particular factors which justified recognizing that prior category in order to determine whether the relationship at issue is, in fact, truly the same as or analogous to that which was previously recognized.”

[31] Among the residual policy concerns that have been held at the second stage of the test to negate a prima facie duty of care is maintenance of the separation of powers as between the executive, legislative, and judicial branches of government. As the Supreme Court stated in Nelson (at para. 43, internal citation omitted), “[i]t is fundamental to the constitutional order that each branch plays its proper role and that ‘no one of them overstep its bounds, that each show proper deference for the legitimate sphere of activity of the other.’”

[32] Reflecting this concern, the Supreme Court has adopted the principle that “core policy decisions” of the legislative and executive branches should be shielded from liability for negligence, as long as they are not irrational or made in bad faith (Nelson at para. 35). Decisions in the “core policy decision” category, the Court has stated (Nelson at para. 44, internal citations omitted),

[33] Here, the motion judge devoted most of her discussion of whether a reasonable cause of action was pleaded in systemic negligence to the question of whether what was pleaded was a breach of a duty of care in policy-making or in the operation of the RCMP in the Territories. She concluded (at paragraph 34 of her reasons) that the material facts pleaded here made it “clear it is operation and not policy at issue,” and that Mr. Nasogaluak was therefore free to allege negligence. In coming to this conclusion, she described the allegations of systemic negligence as directed not to a true policy decision but to “the lack of following policy—thus making it operational,” and drew support from paragraph 51 of the amended statement of claim, which reads as follows:

[34] The motion judge went on to consider whether the operational negligence she concluded was pleaded could be the proper subject of a systemic negligence claim. In determining that it could, she relied on the Supreme Court’s decision in Rumley v. British Columbia, 2001 SCC 69 and that of the Ontario Court of Appeal in Francis v. Ontario, 2021 ONCA 197.

[35] In Rumley, which dealt with a claim by current and former students abused at a residential school for deaf and blind children, the Court (at para. 30) described the systemic negligence claim as alleging a “failure to have in place management and operations procedures that would reasonably have prevented the abuse,” and stated that “[t]hese [were] actions (or omissions) whose reasonability can be determined without reference to the circumstances of any individual class member.” At paragraph 33 of her reasons, the motion judge described this position as “precisely in line with what the Plaintiff is alleging is the case here.”

[36] In Francis, where the Court of Appeal upheld a claim of systemic negligence relating to Ontario’s placement of inmates in administrative segregation, the Court stated (at para. 103) that if it was accepted that actions were taken that made injury to an inmate reasonably foreseeable on an individual basis, there was “no principled reason why that could not be the case on a class basis.” At paragraph 39 of her reasons, the motion judge stated that she saw “strong similarities” between this case and Francis. She also (at paragraph 36) described as “well established” the proposition that governments owe a duty of care to individuals arrested, detained, or otherwise in their custody. This appears to be a reference to a line of cases that includes the Supreme Court’s decision in MacLean v. R., 1972 CanLII 124 (SCC), [1973] S.C.R. 2, referred to further below.

[37] Responding to the Attorney General’s submission that this case was different from other certified systemic negligence claims, which involved “alleged abuse of a specific and identifiable group of individuals under government or institutional care,” the motion judge found that the material facts concerning the relationship pleaded in this case—that “the RCMP is the sole policing agent in the Territories, and that the class members are Indigenous people […] who allege assault by RCMP officers while being held in custody or detained”—met “the proximity and foreseeability test.” Thus, she stated, “the material facts disclose the elements for a duty of care.”

[38] The motion judge concluded her discussion of the systemic negligence cause of action by stating (at paragraph 42) that she “[found] the remaining requisite elements have been pled to show that it is not plain and obvious that this cause of action will fail.”

[39] On appeal to this Court, the Attorney General submits that the motion judge’s analysis of the systemic negligence claim was deficient in three respects: first, in inaccurately characterizing the systemic negligence claim as pleaded as a claim about operation when it is really about core policy; second, in failing to conduct a complete Anns/Cooper duty of care analysis; and third, in failing to consider the viability of the cause of action in negligence as a whole, including the elements of standard of care, breach, causation and damages.

[40] I agree in part with the first of these submissions. As I see it, most of what is pleaded in setting out the systemic negligence claim relates to the operation of the RCMP in the Territories. The declaration sought with respect to negligence, set out above in paragraph 24c, speaks to “the funding, oversight, operation, supervision, control, maintenance, and support” of the RCMP in the Territories. The same or a similar formula is repeated elsewhere in the pleading, in paragraphs 5, 43, 49, 50, and 52. These are at least primarily operational matters. And paragraph 47 pleads that it is as a result of the operation of the RCMP in the Territories that the relationship of proximity with class members arises.

[41] But there are also repeated references in the amended statement of claim to the “establishment” of the RCMP in the Territories as giving rise to liability in systemic negligence, in paragraphs 5 (part of the section headed “overview of this action”), 43, 49, 50, and 52. In my view, the decision to establish the RCMP as the police force in the Territories falls squarely within the definition of a “core policy decision” set out above (at paragraph 32). It is therefore plain and obvious that that decision cannot be the subject of a claim in systemic negligence.

[42] It follows from this conclusion that the “core policy” decision component comprising establishment should be “pruned” from the systemic negligence claim: see Reddock v. Canada (Attorney General), 2019 ONSC 5053 at paras. 409, 442, appeal allowed in part on other grounds, Brazeau v. Canada (Attorney General), 2020 ONCA 184 at para. 115. At this stage of this proceeding, this can most appropriately be accomplished by striking out the portions of the amended statement of claim that plead establishment as part of the systemic negligence claim. I would remit the matter to the motion judge to resolve any issues in light of this discussion as to what if any further portions of the pleading impermissibly advance a claim targeting “core policy” decisions.

[43] I would not give effect to either the Attorney General’s second or third submission with respect to the motion judge’s treatment of the systemic negligence cause of action.

[44] The second, again, is that the motion judge failed to conduct a complete Anns/Cooper duty of care analysis—an analysis that considered first whether the pleading disclosed a relationship of sufficient proximity to give rise to a prima facie duty of care, and then whether any residual policy considerations negated or limited that duty. But although the motion judge did not specifically cite to the Anns/Cooper test, it is clear in my view that she applied all of its relevant components. For example, she expressly determined (at paragraph 41 of her reasons) that the relationship pleaded in the amended statement of claim between the RCMP and Indigenous residents of the Territories who were detained or held in custody was a relationship of proximity and foreseeability. She also addressed the residual policy considerations applicable to “core policy” decisions, considerations that are essentially the same whether they are dealt with at the first or the second stage of the Anns/Cooper framework: Nelson at para. 36.

[45] As I have noted, the motion judge’s conclusion on the existence of a duty of care turned in part on her assessment (at paragraph 36 of her reasons) that it is “well established that governments owe a duty of care to individuals arrested, detained and/or in their custody.” The Attorney General agrees that MacLean is the ultimate source of the duty recognized by the motion judge, but submits that the motion judge failed to appreciate that the duty that flows from MacLean is “entirely distinct” from that alleged here. He submits that MacLean, in which the Crown was found liable in negligence when an inmate suffered serious injuries from a fall at a prison farm, is limited to duties owed by officers to individual prisoners with whom they interact. By contrast, he says, what is alleged here is proximity at the institutional level.

[46] In my view, the scope of the MacLean principle is not yet settled. In Brazeau (at para. 120), the Court of Appeal for Ontario limited the principle to “specific individual acts of negligence,” stating that “[w]hile individual inmates have a cause of action for specific individual acts of negligence on the MacLean principle, a class-wide duty of care can only be made out if the duty relates to the avoidance of the same harm for each class member.”

[47] However, approximately one year later, in Francis (at paras. 102-103), the Court of Appeal saw the principle as having broader application:

[48] The Court added (at para. 110), in a passage quoted by the motion judge here, that “[w]hile individual circumstances may ultimately be relevant to the proof of individual levels of damages, they are not required for proof of a breach of the duty of care on a system-wide basis, nor are they required for determining a base level of damages applicable to all.”

[49] At this screening stage, I conclude that the Attorney General has not shown it to be plain and obvious that there is no reasonable cause of action in systemic negligence based on a failure to plead facts that could establish a duty of care.

[50] That brings me to the Attorney General’s third submission—that the motion judge failed to consider the viability of the cause of action in negligence as a whole, including the elements of standard of care, breach, causation and damages. In light of the discussion above of duty of care, three elements remain to be considered: breach of the duty of care, causation, and damages.

[51] The motion judge’s treatment of these elements is certainly terse. She simply stated (at paragraph 42 of her reasons), “I find the remaining requisite elements have been pled to show that it is not plain and obvious that this cause of action will fail.” However, she had earlier in her reasons (at paragraph 22) written that “[b]reaches of the duty of care include that Canada repeatedly ignored calls for reform and broke promises to do better. The pleadings have the materials [sic] facts supporting that breaches occurred […].” She had also referred to the pleading of material facts supporting the claims of damage.

[52] In any event, it is apparent that three elements are pleaded. Paragraph 60 of the amended statement of claim—reproduced in part above at paragraph 27—sets out a long list of acts and omissions that are said to constitute breaches of the duty of care (as well as breaches of fiduciary duty). Paragraphs 72 to 76 both itemize the injury and damage it is alleged class members have suffered, and attribute that loss and damage to the negligence (and other misconduct) alleged. Contrary to the Attorney General’s submission, the case as pleaded is not a case of negligence “in the air” of the kind decried in Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19 at para. 33.

[53] There is, in my view, no basis for concluding that the Attorney General’s third attack on the motion judge’s treatment of the negligence claim is made out.

[54] The relief sought in paragraph 1 of the amended statement of claim includes

[55] As mentioned above, a declaration is also sought that Canada is liable for damages caused by its breach of fiduciary duty (along with its negligence).

[56] The motion judge began her analysis of the fiduciary duty cause of action by noting Mr. Nasogaluak’s invocation of observations by this Court in Brake v. Canada, 2019 FCA 274 at para. 66. There this Court observed that “the law of fiduciary duty in [the Aboriginal law] context is a very dynamic area of law that is rapidly evolving […],” so that “when the areas of fiduciary obligations and [Indigenous] law intersect…a defendant has a particularly heavy burden in seeking to strike a pleading.”

[57] The motion judge found support for what she characterized as the “novel” claim of breach of fiduciary duty in the Supreme Court’s decision in Manitoba Metis Federation Inc. v. Canada (Attorney General), 2013 SCC 14 at paras. 49-50. There the Court stated that in the Indigenous context, a fiduciary duty on the part of the Crown may arise in two ways. The first is as a result of the Crown assuming “discretionary control over specific Aboriginal interests.” The second, as in private law contexts, is from an express or implied undertaking to “act in the best interests of the alleged beneficiary or beneficiaries.” The Court elaborated that for a fiduciary duty to arise from an undertaking, there must also be “a defined person or class of persons vulnerable to the fiduciary’s control (the beneficiary or beneficiaries),” and “a legal or substantial practical interest of the beneficiary or beneficiaries that stands to be adversely affected by the alleged fiduciary’s exercise of discretion or control.”

[58] The motion judge determined (at paragraph 54 of her reasons) that the conditions set out in Manitoba Metis for a fiduciary duty based on an undertaking to arise were “potentially satisfied” by the material facts pleaded:

[59] She rejected (at paragraph 50) the Attorney General’s submission that because the RCMP’s duties in policing are owed to the public as a whole, the Crown cannot be a fiduciary in relation to the Indigenous people of the Territories. She acknowledged the statement in Alberta v. Elder Advocates of Alberta Society, 2011 SCC 24 at para. 44, that “[t]he Crown’s broad responsibility to act in the public interest means that situations where it is shown to owe a duty of loyalty to a particular person or group will be rare.” But, she stated (at paragraph 53), rarity did not mean impossibility. She also distinguished (at paragraphs 55 and 56) her decision, released the same day as her decision in this case, in BigEagle v. Canada, 2021 FC 504 (appeal heard October 25, 2022). In that case, she declined to certify as inadequately pleaded a claim of breach of fiduciary duty also based on conduct of the RCMP.

[60] The motion judge concluded on this point (at paragraph 57) that while the breach of fiduciary duty cause of action seemed unlikely to succeed, it was not plain and obvious that it would fail.

[61] Before this Court, the Attorney General submits that the motion judge erred in coming to this conclusion. He emphasizes that the RCMP’s “central and public duty is to preserve the peace and to enforce the law for the benefit of all Canadians,” and submits that “[t]he demographics of the Territories, and that there is one police force there, do not convert public duties into fiduciary duties.” He insists, based on Elder Advocates (at para. 54), that “[t]he requirements for fiduciary duty must be rigorously applied […].” He submits (at paragraph 28 of his memorandum) that Mr. Nasogaluak has not pleaded the requisite undertaking, and adds that no viable undertaking could be pleaded here in any event, because “[t]he essential requirement that the fiduciary put the best interests of the beneficiary above all other interests is at odds with the Crown’s—and the RCMP’s—duty to act in the best interests of society as a whole.”

[62] I agree with the Attorney General on this issue. As the Supreme Court further stated in Elder Advocates (at para. 31), “[t]he party asserting the [fiduciary] duty must be able to point to a forsaking by the alleged fiduciary of the interests of all others in favour of those of the beneficiary, in relation to the specific legal interest at stake.” Mr. Nasogaluak does not do so.

[63] While a fiduciary duty may arise from a statute (Elder Advocates at para. 32), the relevant statutory framework here also does not provide for the “forsaking of the interests of all others in favour of those of the beneficiary.” Rather, as the Attorney General submits, it expressly requires RCMP officers to consider the rights of all persons with whom they interact. Section 37 of the Royal Canadian Mounted Police Act, R.S.C. 1985, c. R-10, sets out responsibilities of every member that include

[64] The Code of Conduct of the Royal Canadian Mounted Police, a schedule to the Royal Canadian Mounted Police Regulations, 2014, S.O.R./2014-281, elaborates in calling for members of the RCMP to, among other things,

[65] As the Attorney General submits, it is very difficult to square these duties imposed by statute, which apply in the interactions of members of the RCMP with all those with whom they come into contact, with an undertaking to act in the best interests of the putative class in priority to other interests. In any event, the amended statement of claim pleads no such undertaking. While I am mindful of this Court’s statements in Brake and the Supreme Court’s cautionary statement in Imperial Tobacco (see paragraph 19 above), the Supreme Court has also counselled that “[c]laims against the government that fail to satisfy the legal requirements of a fiduciary duty should not be allowed to proceed in the speculative hope that they may ultimately succeed”: Elder Advocates at para. 54. It is this last proposition, in my view, that applies here.

[66] In my view, the claim of breach of fiduciary duty must be struck out.

[67] Paragraphs 1(d) and (f) of the amended statement of claim set out claims for declarations that “Canada and its agents systematically violated, and continue to violate, sections 7 and 15 of the Charter in a way that is not demonstrably justified in a free and democratic society pursuant to section 1 of the Charter,” and that “Canada is liable to the plaintiff and Class Members for damages under section 24(1) of the Charter for breach of sections 7 and 15 of the Charter in relation to the actions of RCMP Officers.” As noted above, damages of $500 million are sought (in paragraph 1(g)) for negligence and breach of the Charter.

[68] In laying the groundwork for the Charter claims, the amended statement of claim builds (at paragraphs 64 and 65) on the allegations earlier in the pleading (at paragraphs 15, 23 and 24) concerning the conduct of members of the RCMP. This conduct is alleged to include not only arrest, detention, and holding of Aboriginal Persons in custody based on their race, ethnic and /or national origin, but also subjecting those who are arrested or in detention or custody to assaults and excessive force, including beating, hitting, pepper-spraying, and shooting. It is further alleged that the Crown had knowledge of these practices but did nothing about them.

[69] To establish that a law or other government action violates section 7 of the Charter, a claimant must show that the law or action interferes with, or deprives him or her of, life, liberty or security of the person and that the deprivation is not in accordance with the principles of fundamental justice: Ewert v. Canada, 2018 SCC 30 at para. 68.

[70] In setting out the section 7 claim, the amended statement of claim states (in paragraphs 62 to 66) that “[t]he frequency, duration, and severity of the conduct that the Class Members are subjected to at the hands of the Defendant and its agents […] constitutes a breach of the Class Members’ Charter rights to life, liberty and security of the person” and adds that “[t]he widespread use of excessive force […] on Class Members is arbitrary, and grossly disproportionate with the purpose of use of force upon detention,” and is “not carried out in keeping with any principle of fundamental justice.” It is further pleaded that there is no justification under section 1 of the Charter for this excessive use of force.

[71] The motion judge concluded (at paragraph 65 of her reasons) that it was not plain and obvious that the section 7 claim would fail. She noted that the pleading set out the elements necessary to succeed under section 7: government action, a resulting risk to life, liberty, and security of the person, a breach of a principle of fundamental justice, and the absence of justification under section 1. She did not accept the Attorney General’s submission, based on the decision of the Court of Appeal for Ontario in JB v. Ontario (Child and Youth Services), 2020 ONCA 198, leave to appeal refused, 39165 (October 8, 2020), that the section 7 claim should not be certified because it was merely a recasting of the negligence claim. She distinguished the JB case on the facts and on the basis that the claim here is a “top-down” claim—one not grounded in individual circumstances.

[72] The motion judge also rejected the Attorney General’s further submission that the section 7 claim (and the section 15 claim) should not proceed because some of the conduct in question would have preceded the coming into force of the Charter. While this might cause some class members’ claims to fail at the individual stage, it did not, she stated, mean that the cause of action as a whole was doomed to fail.

[73] On appeal, the Attorney General’s principal submission is that the motion judge erred in failing to reconcile a “fundamental inconsistency” in the claim, treating it as a “top-down” claim but analyzing the cause of action as a series of individual breaches, with no particular systemic conduct. He submits that if the section 7 claim is truly “top-down,” the conduct of particular RCMP members would not be relevant, and the Charter claim would amount to an impermissible re-pleading of the negligence claim.

[74] I would not give effect to this submission. As the motion judge concluded, all of the elements of a section 7 cause of action are pleaded. The government action pleaded in support of the section 7 claim includes the systemic excessive use of force. While similar conduct is also pleaded in support of the systemic negligence claim, that is not by itself a basis for finding that there is no reasonable section 7 cause of action; see, for example, Francis at paras. 94-110, where liability was found both in negligence and under section 7 (and section 12) of the Charter for the same underlying conduct. The fact that the claim is a “top-down” claim does not mean that the conduct alleged in the operation and management of the RCMP could not have resulted in breaches of individual class members’ Charter rights. Even if there cannot be liability both in negligence and under the Charter for the same conduct, I would leave it to the common issues trial judge to determine which (if either) of the two categories of claims should succeed.

[75] I see no reversible error in the motion judge’s assessment that it is not plain and obvious that the section 7 claim will fail.

[76] Section 15(1) of the Charter states:

[77] To prove a violation of section 15(1), a claimant must show that the impugned law or state action (1) creates a distinction based on enumerated or analogous grounds, on its face or in its impact, and (2) imposes a burden or denies a benefit in a manner that has the effect of reinforcing, perpetuating, or exacerbating disadvantage: R. v. Sharma, 2022 SCC 39 at para. 28; Fraser v. Canada (Attorney General), 2020 SCC 28 at para. 27.

[78] As will be apparent from paragraphs 67 and 72 above, the section 15(1) claim is in large measure pleaded in tandem with the section 7 claim. In the portion of the amended statement of claim specifically directed to the breach of section 15(1) (paragraphs 67 to 70), it is alleged that class members have been discriminated against based on, among other things, their race and their national, spiritual, religious and ethnic origin, and thus treated them negatively based on enumerated grounds. The discriminatory actions are said to include the RCMP (1) allowing its agents to target Aboriginal Persons during their time in custody, (2) allowing its agents to use excessive force on Aboriginal Persons in custody, and (3) acting carelessly, recklessly or wilfully blindly, or deliberately accepting or actively promoting, a policy of discrimination against Aboriginal Persons in custody in the Territories. There is no section 1 justification, it is pleaded, for this conduct.

[79] Before the motion judge, the Attorney General conceded that the first two of these allegations, though not the third, met the test for a reasonable cause of action. However, as he did in relation to the section 7 claim, he raised the objection that the section 15(1) claim could nonetheless not proceed because it included a period that preceded the coming into force of the Charter’s equality guarantee.

[80] The motion judge saw this issue as one going to the “common issues” condition for certification, and stated that if it posed a problem, the problem could readily be addressed by dividing the class into two subclasses. The section 15(1) claim accordingly satisfied the reasonable cause of action condition.

[81] The Attorney General does not challenge this determination in this Court.

[82] For the reasons set out above, I conclude that the first certification condition is met, but only in relation to the systemic negligence claim (as pruned) and the Charter claims.

[83] For ease of reference, I repeat the proposed class definition:

[84] For the “identifiable class” condition to be met, “the evidence must support some basis in fact for an objective class definition that bears a rational connection to the litigation that is not dependent on the outcome of the litigation”: Greenwood at para. 168, citing authorities in this Court and the Supreme Court of Canada.

[85] The motion judge found these criteria to be met. She did not accept the Attorney General’s submissions that the proposed definition was imprecise, overbroad, and unmanageable, or, more specifically, (1) that it was based on non-objective criteria because the criterion of having made an allegation was not sufficiently objective; (2) that it was unnecessarily complex because it would require determining whether proposed class members would be considered Aboriginal under section 35 of the Constitution Act, 1982; (3) that it was overbroad because the proposed class period dated back to 1928; (4) that it was over-inclusive because its members would have had disparate experiences involving the RCMP; (5) that it was unmanageable because it comprised individuals who have merely made allegations; (6) that it lacked any basis in fact because the affidavit evidence did not support allegations of unlawful assaults; and (7) that it had no rational connection to the proposed common issues because there was no causal link pleaded between the assault allegations and the common issues.

[86] In rejecting these submissions, the motion judge noted that a class definition based on claiming to have suffered injury, loss, or damage as a result of the alleged misconduct was accepted in Rumley. She also expressed the view that determining Aboriginal status would not be overly complex.

[87] Before this Court, the Attorney General maintains substantially the same objections to the proposed class definition as he made before the motion judge. He submits that the class definition as framed—based on allegations of assault—is not objective but subjective. He points out that assault requires intention on the part of the perpetrator to cause the victim to reasonably apprehend imminent harm: McLean v. McLean, 2019 SKCA 15 at para. 59. He adds that section 25 of the Criminal Code, R.S.C. 1985, c. C-46, authorizes police officers to use reasonable force, thus further complicating the question whether allegations of assault involve true assaults.

[88] The Attorney General goes on to submit that Rumley is distinguishable: it involved allegations by members of a narrowly defined group and a clear duty on the part of the province to protect them. He points out that claims-based class definitions are not universally accepted and submits that, where they are accepted, the claims are linked to objective criteria. He argues that a rational connection between the class definition and the common issues is absent in this case, and that absence is fatal to a class proceeding. He also submits that, in any event, there is no support in the evidence for a class period that dates back to 1928, because the affidavit evidence of the members of the putative class describing mistreatment by the RCMP goes back only to 1990.

[89] In response, Mr. Nasogaluak submits that the class definition is objective rather than subjective, and that the motion judge was entitled to rely on the similarity to Rumley in approving the definition here. He provides a non-exhaustive list of six other class proceedings that have been certified with claims-based class definitions. He refers this Court to the comments by Winkler J. (as he then was) in Attis v. Canada, 2007 CanLII 15231 (ON SC) at paras. 56-57, addressing the criticisms of claims-based class definitions, and expressing the view that, when utilized in appropriate cases, they are fully consistent with the purposes of class definition. Those purposes, as stated by the Supreme Court in Western Canadian Shopping Centres Inc. v. Dutton, 2001 SCC 46 at para. 38, are to identify the individuals entitled to notice, entitled to relief (if relief is awarded), and bound by the judgment.

[90] It does not appear this Court has considered the appropriateness or inappropriateness of claims-based class definitions. However, other appellate courts have done so. In Merck Frosst Canada Ltd. v. Wuttunee, 2009 SKCA 43, leave to appeal refused, [2008] S.C.C.A. No. 512, the Court of Appeal for Saskatchewan reviewed a number of certification decisions in other provinces (including Attis), some accepting and others rejecting claims-based class definitions. It concluded as follows: (para. 103):

[91] The Court proceeded to accept (at para. 106, emphasis in original) that “a definition in terms of those who claim loss or injury […] would itself be related to an objective, verifiable fact or event […].”

[92] The Saskatchewan Court’s statement in Wuttunee, quoted above in paragraph 90, was approved by the Court of Appeal of Newfoundland and Labrador in Dow Chemical Company v. Ring, Sr., 2010 NLCA 20 at para. 71, leave to appeal refused 2010 CanLII 61130 (SCC). Its approach to the identifiable class issue was also applied by the Court of Appeal for British Columbia in Kwicksutaineuk/Ah-Kwa-Mish First Nation v. Canada (Attorney General), 2012 BCCA 193 at paras. 90-99, leave to appeal refused, 2012 CanLII 70221 (SCC).

[93] I too would adopt the approach in Wuttunee. In my view, the claim-based class definition applied in this case is sufficiently objective having regard to the purposes of defining the class.

[94] I should perhaps note that in coming to this conclusion, I do not rely heavily on Rumley. As others have pointed out, it does not appear that the class definition was contested by the time the Rumley case reached the Supreme Court, and the Court made no comment concerning the issue: see Wuttunee at paras. 97, 102, quoting from Attis at para. 55.

[95] I turn now to the Attorney General’s contention that the motion judge committed a palpable and overriding error by accepting the proposed class definition despite the absence of the necessary rational connection between the class definition and the common issues. I do not accept this contention. As Mr. Nasogaluak submits, this argument fails to recognize the systemic, “top-down” theory of his case. According to the Attorney General, the individual assaults must first be proven before any inquiry into the alleged systemic failings can take place. But what is alleged on behalf of the putative class is that the individual harm arose out of the systemic failings identified in the amended statement of claim. As the motion judge recognized (at paragraph 102 of her reasons),

[96] As for the class period, it was an agreed fact before the motion judge (noted at paragraph 6 of her reasons) that the RCMP has been conducting policing in the Territories since 1928. The record before her included an acknowledgment by the current Commissioner of the RCMP not only that systemic racism now exists within the RCMP, but also that “[t]hroughout [its] history and today, [it has] not always treated racialized and indigenous people fairly”: Statement by Commissioner Brenda Lucki dated June 12, 2020, Appeal Book, p. 3138. This acknowledgment meets the requirement of some basis in fact to support the class period.

[97] While the manageability of the class as defined will likely present some challenges, given the length of the class period and the disparate locations potentially involved, techniques exist to address them: see Rumley at paras. 31-32. Furthermore, the inclusion in the class definition of the proviso that class members must have been alive as of December 18, 2016 should go some way to limit the complexity.

[98] Finally with respect to this condition, I do not accept that use of the term “Aboriginal persons” gives rise to insurmountable complexity in identifying members of the class. At paragraph 2(a) of the amended statement of claim, the definition of the terms “Aboriginal” and “Aboriginal person(s)” is directly linked to the holding of rights under section 35 of the Constitution Act, 1982. The Federal Court approved a similar definition in granting certification in Gottfriedson v. Canada, 2015 FC 766 and 2015 FC 706. In most cases it can be expected to be clear who comes within the definition. If disputes arise, judicial guidance as to the application of the definition is available: see, for example, R. v. Desautel, 2021 SCC 17.

[99] I see no reversible error in the motion judge’s conclusion on the “identifiable class” issue.

[100] In Greenwood, this Court described the nature of the inquiry into the satisfaction of the “common issues” condition in the following terms (at para. 180, citing Supreme Court authorities and this Court’s decision in Brake):

[101] The motion judge certified the following questions as common issues.

[102] In certifying these common issues, the motion judge found that they provided an appropriate means of moving the litigation forward, avoiding duplication of legal analysis. She did not accept the Attorney General’s submission—one repeated before this Court—that an inordinate amount of individual fact-finding would be necessary. Rather, she stated (at paragraph 101 of her reasons), “[a]nswering the common questions in either the positive or the negative will then allow for the determination of the individual class members [sic] situations where some degree of individual assessment will be appropriate.” That is precisely what the litigation plan approved by the motion judge provides for. Of course, if the common questions going to liability are answered in the negative, no individual fact-finding may be necessary at all.

[103] The Attorney General contends that the motion judge’s further statement (at paragraph 102 of her reasons) that “damage to the class member is both evidence of the system as well as potential cause for damages” makes it clear that determination of the common issues is dependent on, and must necessarily follow, individual findings of fact. I do not agree: I see the statement merely as another recognition of the systemic nature of Mr. Nasogaluak’s claim.

[104] The Attorney General further submits that the certified common issues go beyond the scope of the evidence before the motion judge. But as Mr. Nasogaluak responds, the evidence on a certification motion need not establish the case on the merits; it need only show some basis in fact for the common issues: Pro-Sys at paras. 100, 110.

[105] Here, the record included, in addition to the affidavit evidence referred to at paragraphs 15 and 16 above, and the acknowledgment of Commissioner Lucki referred to at paragraph 96, expert affidavit evidence of Scot Wortley, Ph.D., Associate Professor at the Centre of Criminology and Sociolegal Studies, University of Toronto. He deposed that he would analyze documents available through the discovery process, using models that he described, and that this analysis would help him determine the extent to which racial bias and systemic failures contribute to RCMP use of force against Indigenous peoples in the Territories, and whether RCMP data collection, use of force policies, anti-bias policies and training are effective, appropriate, and meet the standards required to ensure equitable policing in the Territories: Appeal Book, PDF pp. 2894, 3149.

[106] As the express language of the condition in rule 334.16(1)(c) makes clear, the “common questions” condition for certification may be met “whether or not those common questions predominate over questions affecting only individual members.” Numerous systemic claims similar to those here have been found to meet this condition or similarly worded conditions in many other cases: see, for example, Rumley at paras. 27, 30; Canada v. John Doe, 2016 FCA 191 at para. 63; Greenwood at paras. 183-184; Francis at paras. 106-107. Many further examples are provided at footnote 103 of the respondent’s memorandum.

[107] With two exceptions, I agree with the motion judge that there is some basis in fact for the common questions she certified, and that she committed no palpable and overriding error in doing so.

[108] The first exception flows from my conclusion that the claim for breach of fiduciary duty cannot be certified. It follows that the question directed to this cause of action—question (b)—should be set aside.

[109] The second exception relates to the motion judge’s certification of a question—question (h)— concerning aggregate damages. It reads as follows:

[110] Rule 334.28(1) authorizes a judge to make any order in respect of the assessment of monetary relief, including aggregate assessments, that is due to the class or a subclass. However, the record before the motion judge contained no evidence of any method for conducting an aggregate assessment, and the litigation plan similarly says nothing about how an aggregate assessment would be conducted.

[111] In Greenwood, in the face of a similar factual vacuum, this Court found (at paras. 188-189) that there was no basis in fact for certifying a common question related to the assessment of aggregate damages, and that the motion judge had committed a palpable and overriding error in doing so. Similarly, in Lewis v. WestJet Airlines Ltd., 2022 BCCA 145 at paras. 158-160, leave to appeal refused, 2019 CanLII 64825 (SCC), the Court of Appeal for British Columbia recently upheld a decision of a certification judge declining to certify a common issue on aggregate damages in comparable circumstances.

[112] In my view, it is appropriate that we follow Greenwood on this issue.

[113] Mr. Nasogaluak submits that we should not do so, because the aggregate damages claim here includes a claim for Charter damages, a “flexible” category of damages not at issue in Greenwood. He further submits that Dr. Wortley’s reports can serve to provide a methodology for assessing Charter damages.

[114] I disagree. The fact that the aggregate damages claimed here consist in part of Charter damages does not relieve Mr. Nasogaluak of the obligation to put forward a methodology for determining them. And I see nothing in Dr. Wortley’s reports that is directed to fulfilling, or would fulfil, that obligation.

[115] I would accordingly require an amendment to the certification order to delete the common question on aggregate damages. Its deletion would not preclude the common issues trial judge from concluding that the requirements for an aggregate award of damages are met and from making an aggregate award, if appropriate, without individual assessments: Pro‑sys at para. 134; Lewis at para. 160; Good v. Toronto Police Services Board, 2016 ONCA 250 at para. 75. Thus, despite my conclusion concerning the aggregate damages claim at this stage of the proceeding, the possibility remains that damages may be awarded on an aggregate basis, at the common issues trial, and that not solely through subsequent individual assessments.

[116] As the motion judge stated (at paragraph 113 of her reasons), citing Hollick at para. 27 and AIC Limited v. Fischer, 2013 SCC 69 at para. 22, the preferability inquiry is to be conducted through the lens of the three main goals of class proceedings: judicial economy, behaviour modification, and access to justice. She found that “[w]hen looked at through this lens, and taking into account the vulnerable state of northern Indigenous people, it is evident that a class action is the preferable procedure for the resolution of [the common] issues.”

[117] The motion judge did not accept the Attorney General’s submissions that individual civil actions or some form of complaints procedure would be a better means of redress for those alleging assault, and that the systemic issues could better be addressed though a public inquiry. She found that the general economic status of class members would make individual actions prohibitive. There was, in any event, no evidence to indicate that either a complaints procedure or a public inquiry would be forthcoming.

[118] The motion judge also considered and applied rule 334.18(a), which states that a judge

[119] Before this Court, the Attorney General maintains that the motion judge committed a palpable and overriding error in finding that a class action is the preferable procedure. He submits that the motion judge was required by both rule 334.16(2) and Hollick to consider the importance of the common issues in relation to the claims as a whole, but failed to do so.

[120] I do not agree that there was any failure as alleged. In paragraph 109 of her reasons, in the portion of the reasons addressing preferable procedure, the motion judge expressly referred to the Attorney General’s submission that

[121] She added that the Attorney General had “mentioned that even if a duty of care were to be found, it would leave breach of the standard of care, causation, and damages to be individually assessed.”

[122] While the motion judge did not accept the Attorney General’s position on this point, she plainly considered the importance of the common issues in relation to the claims as a whole in deciding that a class action was the preferable procedure. I see no palpable and overriding error in that determination—a determination entitled to deference.

[123] I would allow the appeal in part, set aside the certification order, and remit the order to the Federal Court to be amended in accordance with these reasons. The Federal Court should also require that the litigation plan be amended to conform with these reasons. Applying rule 334.39, I would make no order as to costs.