Dunsmuir and Jurisdiction (Evan Fox-Decent and Alexander Pless)

Evan Fox-Decent is a Full Professor at the Faculty of Law, McGill University and Alexander Pless is General Counsel at the Department of Justice (and his views are expressed in a personal capacity)

In 1979, two Supreme Court of Canada decisions lay the groundwork for a new era of judicial review. In Nicholson v Haldimand-Norfolk Regional Police Commissioners,[1] Laskin CJ, for a 5-4 majority, affirmed that a probationary constable was owed a common law duty of fairness; he was entitled to be heard prior to a decision to terminate his services. Prior to Nicholson, the constable would have been subject to termination without a hearing and without disclosure of the reasons for his dismissal. In CUPE v NB Liquor Corporation,[2] Dickson J (as he then was), for a unanimous Court, sought to rein in judicial interventionism in labour law by admonishing reviewing courts to “not be alert to brand as jurisdic­tional, and therefore subject to broader curial review, that which may be doubtfully so.”[3] Dunsmuir v New Brunswick[4] arguably marks a significant backslide from Nicholson and CUPE. Under Dunsmuir, probationary constables are not owed a common law duty of procedural fairness, and jurisdiction acquires new salience as a basis for substantive review.

In this contribution, we focus on one set of the implications from Dunsmuir with respect to substantive review and CUPE’s legacy. We suggest that the Court’s discussion in Dunsmuir of jurisdictional review is in deep tension with the deferential approach Dickson J inaugurated with CUPE, and that eventually became known as the pragmatic and functional approach. At the root of the tension is the Court’s failure—both before and after Dunsmuir—to provide a basis for distinguishing jurisdictional from non-jurisdictional errors of law, a failure that risks inviting courts to ‘brand as jurisdictional…that which may be doubtfully so.’

CUPE itself supplies an instructive example of the importance of the jurisdictional/non-jurisdictional distinction. At issue in the case was the meaning of the term “employee” within the Public Services Labour Relations Act,[5] and more specifically, whether management personnel were to count as “employees” for the purpose of determining whether they could lawfully do the work of striking employees. Interpretative arguments pulled in both directions. Management argued that “employee” was a defined term under the Act. The union argued that the Act prohibited management personnel from replacing striking workers as the quid pro quo for the Act’s prohibition of union picketing.[6] The lower courts held that determining the meaning of “employee” was a matter of legal interpretation and jurisdictional, and therefore the Board had to get the interpretation right to have jurisdiction to determine whether management personnel were unlawfully replacing striking workers. Having thus determined the crucial issue to be jurisdictional in nature, the lower courts reviewed the Board’s decision on a standard of correctness, ultimately quashing it.

Dickson J criticized this jurisdictional approach. He pointed out that the lower courts failed to acknowledge the legislature’s decision to confide these decisions to an administrative body, and likewise failed to consider the good reasons the legislature had to protect the decisions of the Board with a privative clause (e.g., specialized expertise, considerations of subsidiarity). For Dickson J, these considerations suggested that the Board was owed deference, and that its decisions would be set aside only if they were found to be patently unreasonable. Since the legislation admitted more than one reasonable interpretation, including the quid pro quo interpretation defended by the Board, Dickson J held that the appropriate course for the courts was to defer to the Board. Dickson J did not, however, declare a ban on jurisdictional review, and of equal note, he did not explain how courts were to determine if a question of legal interpretation was or was not jurisdictional.

In the intervening years between CUPE and Dunsmuir, jurisdictional review would sometimes make an appearance, though not with the ascendancy it enjoyed prior to CUPE.[7] Stained with the mark of judicial interventionism, by 1998 jurisdictional review had retreated to such an extent that Bastarache J could declare that it was no more than an alternative label appropriate for those cases where application of the contextual factors of the pragmatic and functional approach determined that correctness was the proper standard of review.[8] But like Dickson J in CUPE, rather than offer a principled means of distinguishing jurisdictional issues from others, Bastarache J’s strategy was simply to attempt to marginalize jurisdictional review.

In Dunsmuir, the chickens come home to roost. Bastarache and LeBel JJ unabashedly write that “[i]n essence, the rule of law is maintained because the courts have the last word on jurisdiction.”[9] They underline the role of jurisdictional review by citing Crevier v AG (Québec) et al[10] and U.E.S., Local 298 v. Bibeault, and characterize the “inherent power of superior courts” as a power “to review administrative action and ensure that it does not exceed its jurisdiction.”[11] Regarding private clauses, the majority writers opine that “[j]udicial review is necessary to ensure that the privative clause is read in its appropriate statutory context and that administrative bodies do not exceed their jurisdiction.”[12] And in their discussion of correctness, the two judges insist that they are not calling for a return to jurisdictional review as it was practiced pre-CUPE, but nonetheless affirm that a standard of correctness applies to “true questions of jurisdiction or vires.”[13]

Bastarache and LeBel JJ insist that “jurisdiction” is to be understood “in the narrow sense of whether or not the tribunal had the authority to make the inquiry.”[14] They also say, apparently as a gloss on their ‘narrow sense’ idea, that “[i]n other words, true jurisdiction questions arise where the tribunal must explicitly determine whether its statutory grant of power gives it authority to decide a particular matter.”[15] This latter idea, however, is not a gloss on the first. The first idea refers to the scope of the tribunal’s statutory power. The second idea refers to a condition that, if satisfied, denotes that a jurisdictional issue is in play (i.e., a tribunal being compelled to determine the scope of its own power), but says nothing about what the scope of a tribunal’s authority might be nor how questions of scope are to be determined.

The majority insists that jurisdictional review must be based on “narrow” questions, offering United Taxi Drivers’ Fellowship of Southern Alberta v. Calgary (City)[16] as an example. In that case, the Court had to determine whether the City of Calgary had lawful authority to enact bylaws limiting the number of taxi licenses. The difficulty with this reasoning is that any question of law can be brought to fit (narrowly!) within it. For example, did the Board in CUPE have legal authority to prohibit management personnel from replacing striking workers? Or, did the labour adjudicator in Dunsmuir have authority to inquire into the actual reasons for Dunsmuir’s dismissal? These question are both narrow and evocative of jurisdictional concerns. But if Dunsmuir permits or requires this approach to review, then we are perilously close to the kind of inquiry Dickson J was anxious to avoid in 1979.

In Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association,[17] Rothstein J. attempted to alter course, suggesting that “it may be that the time has come to reconsider whether, for purposes of judicial review, the category of true questions of jurisdiction exists and is necessary to identifying the appropriate standard of review.”[18] He pointed out the inherent contradiction between Dunsmuir’s presumption that deference is owed to an administrative body interpreting its home statute, and an application of the correctness standard for so-called “true questions” of jurisdiction, since the latter ordinarily involves the former.  Despite some recent affirmation of Rothstein’s obiter dictum,[19] Dunsmuir’s jurisdictional questions remain.  Indeed, in their dissenting reasons in Quebec (Attorney General) v Guérin,[20] Justices Brown and Rowe explicitly disavowed Rothstein’s plea in favor of maintaining the category of “true questions” of jurisdiction, questions that would always be reviewed on a standard of correctness.

The Court in Dunsmuir claims it is following CUPE. Its reanimation of jurisdictional review, however, is in deep tension with CUPE’s message of deference

[1] Nicholson v Haldimand-Norfolk Regional Police Commissioners, [1979] 1 SCR 311 [Nicholson].

[2] CUPE v NB Liquor Corporation, [1979] 2 SCR 227 [CUPE].

[3] Ibid., at p.233.

[4] Dunsmuir v New Brunswick, [2008] 1 SCR 190 [Dunsmuir].

[5] Public Service Labour Relations Act, R.S.N.B. 1973, c. P‑25, [the Act].

[6] CUPE at p.240.

[7] See e.g., U.E.S., Local 298 v. Bibeault, [1988] 2 SCR 1048 per Beetz at paras 106-127.

[8] Pushpanathan v Canada (Minister of Citizenship and Immigration), [1998] 1 SCR 982. The contextual factors were (1) presence of a private clause or statutory right of appeal; (2) relative expertise; (3) nature and role of the statutory provision; and, (4) nature of the issue (law, mixed fact and law, fact). Ibid. at paras 29 to 38.

[9] Dunsmuir, supra note 4 at para 30.

[10] Crevier v AG (Québec) et al [1981] 2 SCR 220.

[11] Dunsmuir, supra note 4 at para 31.

[12] Ibid. at para 52.

[13] Ibid. at para 59.

[14] Ibid.

[15] Ibid.

[16] United Taxi Drivers’ Fellowship of Southern Alberta v. Calgary (City) 2004 SCC 19.

[17] Alberta (Information and Privacy Commissioner) v Alberta Teachers’ Association, 2011 SCC 61.

[18] Ibid at 34.

[19] Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd., 2016 SCC 47 at para 26.

[20] Quebec (Attorney General) v Guérin  2017 SCC 42.

This content has been updated on February 25, 2018 at 23:30.