In the common law tradition, courts are at the apex of the interpretive hierarchy. Traditionally, their role has been to answer legal questions authoritatively. When inferior bodies, including administrative decision-makers,[i]commit jurisdictional errors the superior courts will intervene to correct them.

A deferential approach to judicial review requires judges, however, to be satisfied by an answer that is merely reasonable, even on questions of law.[ii]Deferring to administrative decision-makers’ interpretations of law requires judges to pull against the current of tradition. [iii]Intervention is justifiable only in extreme cases, not in ordinary ones. Administrative autonomy must be respected, tradition put to one side. A deferential approach requires judges to no longer think as lawyers traditionally have thought.

Since the 1970s, the Supreme Court of Canada has attempted to chart a deferential course to judicial review of interpretations of law. It has repeatedly been buffeted by traditionalist storms pushing it towards a more interventionist course. The concept of jurisdictional error, with its teaching that on some matters an administrative decision-maker must be correct or face judicial intervention, has usually been at the eye of the storm.[iv]

Deference can take several forms, one of which is the creation of exclusive spheres of authority.[vi] During judicial review’s twilight years, administrative decision-makers were free to make errors within jurisdiction. They had the ‘right to be wrong’.[vii] Twilight evokes darkness and in cases decided before the modern law of judicial review began to take shape courts were indeed unwilling to shine a light on certain areas of administrative decision-making.

However, jurisdictional limits were policed by the courts, sometimes jealously.[viii]In addition to its strong position on errors of law, the Supreme Court of Canada took an aggressive stance against erroneous findings of jurisdictional fact.[ix] Although the distinction between jurisdictional and non-jurisdictional error accorded deference within limits, it became clear that those limits would be so strictly policed by reviewing courts that deference was largely illusory. No real space was carved out for administrative decision-makers, whose autonomy ultimately depended on the good grace of the judges. Underneath the technical language of jurisdiction and error of law ran a current of hostility to administrative bodies.

And that technical language was capable of manipulation to achieve the end of undermining administrative decision-makers. The distinction between jurisdictional and non-jurisdictional errors of law and fact was “sufficiently vague – some would say meaningless – to offer a means of review of any erroneous finding of fact or law”.[x] Accordingly, a broad conception of jurisdictional error allowed reviewing courts to reduce the autonomy of administrative decision-makers and keep them within closely confined boundaries. Often, the Court would be “utterly submerged by a mistaken zeal to ‘correct’ the agency’s decision”.[xi] Commenting later on Anisminic and Metropolitan Life, Wilson J suggested that they reflected a failure to appreciate “(1) that [administrative decision-makers’] decisions are crafted by those with specialized knowledge of the subject matter before them; and (2) that there is value in limiting the extent to which their decisions may be frustrated through an expansive judicial review”.[xii]

A discussion of CUPE follows…

Where the legislature had created a specialized, expert tribunal and protected its decisions by way of a privative clause, courts could intervene only in extreme cases:

The composition and institutional structure of the agencies, together with the expertise and the wide range of procedural tools available to them, apparently persuaded the courts that these bodies had indeed been given the primary statutory responsibility for implementing and elaborating the legislative mandate within their area of regulation.[xiii]

This was the case in many fields other than labour law, which dominated the Court’s development of a deferential approach.

Despite the eloquence of Dickson J’s formulation, the concept of jurisdictional error remained afloat. Dickson J’s decision in CUPE left the jurisdictional door ajar, though with a warning not to go through it. All Dickson J had done was to urge caution on the part of reviewing courts, an attitudinal shift. It remained true that “[n]o satisfactory test has ever been formulated for distinguishing findings which go to jurisdiction from findings which go to the merits”.[xiv] Where a question could be dressed in jurisdictional garb, intervention by a reviewing court would remain justifiable. And given the hold exerted over the lawyerly mind by traditionalist conceptions of jurisdictional error and the rule of law, the attitudinal shift urged by Dickson J would be insufficient to withstand interventionist tides.[xv]

It was necessary to formulate a test for determining whether or not a question was jurisdictional in nature. In Union des employés de service, local 298 v Bibeault,[xvi]  Beetz J found a solution in the Court’s growing acceptance of the relative institutional competence of administrative decision-makers. The formal and substantive reasons for deference provided a way to answer the key question of legislative intent: did the legislature intend that the decision-maker, or the reviewing court, should take the decision at issue? To answer this question, Beetz J held, a reviewing court should look at “the wording of the enactment conferring jurisdiction on the administrative tribunal…the purpose of the statute creating the tribunal, the reason for its existence, the area of expertise of its members and the nature of the problem before the tribunal”.[xvii] Formal and substantive reasons for deference would be revealed by this analysis.

This “pragmatic and functional” analysis reached its logical conclusion in Pushpanathan v Canada (Minister of Citizenship and Immigration).[xviii] Bastarache J outlined a multi-factor test designed to require justification for judicial intervention in terms of legislative intent, “a bold judicial venture designed to infuse jurisdiction with meaningful normative content”.[xix] Only those questions the legislature intended to reserve to the courts should be treated as “jurisdictional”, to be determined by reference to four factors: the presence or absence of a privative clause or right of appeal;[xx] the relative expertise of the decision-maker; the purpose of the statutory framework; and the nature of the question at issue.

Jurisdictional error as a stand-alone concept was gutted.[xxi] “Jurisdiction” became“simply descriptive of a provision for which the proper standard of review is correctness, based upon the outcome of the pragmatic and functional analysis”:

In other words, “jurisdictional error” is simply an error on an issue with respect to which, according to the outcome of the pragmatic and functional analysis, the tribunal must make a correct interpretation and to which no deference will be shown.[xxii]

To say that a statutory provision was “jurisdictional” and required judicial review on a standard of “correctness” was simply to say that, on an application of the pragmatic and functional analysis, the question was one for the courts, not for the administrative decision-maker. Moreover, the “nature of the question” was just one factor amongst several. Categorizing a question as “legal” would not be sufficient to justify judicial intervention. Deference on questions of law seemed firmly anchored and sheltered by a multi-factor test from the rip-tide of jurisdictional error.

Despite the apparent triumph of the pragmatic and functional analysis, interventionist tides continued to drag judges back towards more familiar shores: “The temptation for judges to slip into old ways of thinking in which the formal concept of jurisdiction is an active legal norm rather than a passive legal outcome has sometimes been overwhelming”.[xxiii] Questions that would traditionally have been seen as jurisdictional had an almost magical capacity to attract judicial intervention, the multi-factor test notwithstanding; “jurisdictional error began to resurface but in a shadowy form”.[xxiv] Indeed, when the Court revised the judicial review framework in Dunsmuir v New Brunswick, “true” jurisdictional questions were said to attract a standard of review of correctness,[xxv] a tacit admission of the persistence of the concept of jurisdictional error standing alone.[xxvi]

The reader will not be surprised to learn that appeals often attracted the application of the correctness standard of review, at least for “pure questions of law”,[xxvii]especially where these had “significant precedential value”.[xxviii]Thus in Pushpanathan v Canada (Minister of Citizenship and Immigration), which would otherwise represent the apotheosis of the pragmatic and functional analysis, a question of international law was held to require the application of a standard of correctness.[xxix]Unsurprisingly, in Dunsmuir, the Court retained a category of questions of general law of central importance to the legal system which were to be answered by courts.[xxx]  

Implementing deference proved very difficult. The failure to do so satisfactorily led to reformulation of judicial review doctrine in Dunsmuir v New Brunswick,[xxxi] which in turn set the scene for the modern equivalent of previous battles for deference. Today the question is whether a revised conception of reasonableness will withstand the traditional interventionist forces of jurisdictional error and error of law.

The second storm is potentially much more damaging. It involves the second prong of the reasonableness standard in Dunsmuir: that decisions must fall within a “range of possible, acceptable outcomes”. In determining the outer limits of the “range” it appears that the reasonableness of administrative interpretations of law is adjudged by reference to judicial principles of statutory interpretation, such that little separates modern reasonableness review from prior interventionist approaches.

Yet a willingness to assert judicial supremacy by using the principles of statutory interpretation to set the limits of the “range” of acceptable interpretive outcomes, or by identifying “clear” statutory provisions further underlines the power of traditionalist thinking. The very purpose of the principles of statutory interpretation is to help judges to find the best answers to legal questions. They are not designed to identify ambiguity; they are designed to resolve it. This approach requires courts to insist on the application of judiciallycrafted principles of interpretation to circumscribe the authority of administrative decision-makers, holding non-lawyers to legal standards and moving them to adopt an interpretive mindset that is not necessarily their own. Administrative decision-makers are taught by reviewing courts that it is better to think like lawyers, even if the legislature’s goal of empowering administrative decision-makers in the first place was to avoid having decisions made by people who think like lawyers.[xxxix]

In addition, declaring that “clear” provisions must be enforced by reviewing courts is very much like saying that “jurisdictional” provisions must be enforced by reviewing courts. The distinction between “clear” and “unclear” statutory provisions threatens to prove just as tenuous and malleable as the distinction between jurisdictional and non-jurisdictional error.[xl] Both distinctions have their origins deep in traditionalist waters.

The emergence of deference in Canadian administrative law was contingent on several factors. That the emergence of deference may well have been an accident of history would explain much of the doctrinal “ebb and flow” that Canadian administrative law has experienced over the past forty years.[l] An attitude championed by a cadre of true believers in deference must inevitably struggle for dominance with traditionalist common law thinking. It remains necessary “to provide antidotes to the propensity of courts to see themselves as experts on many issues of regulatory law and to downplay the capacities of tribunals in relation to those same questions”.[li] It is questionable whether the emerging presumption of reasonableness review will provide the necessary antidotes. For interventionist judges will more readily find clear answers to questions of statutory interpretation or narrowly confine the range of permissible decisions and they will no longer have to couch these conclusions in terms of “the general context and operation of a statute before applying it”.[lii]Requiring a cursory nod to the presumption of reasonableness review, unaccompanied by an appreciation of the relative institutional competence of the decision-maker, is unlikely to cause judges to go against the common law’s traditional flow.[liii]

The problem may well be what David Dyzenhaus termed the “paradox of the recognition of rationality”: “To recognise rationality is at the same time to claim a judicial role in supervising the administrative process to ensure that it meets standards of rationality, even if a sincere attempt is made to conceive those differently”.[liv] There is no doubting the sincerity of Canadian judges, past and present. But there is no doubting either that they have struggled to swim against the current. Given deference’s apparently unending struggle against tides of tradition and interventionism the closing words of F Scott Fitzgerald’s The Great Gatsby seem apt: So we beat on, boats against the current, borne back ceaselessly into the past.

---