Legal Form and Administrative Discretion

Lessons from Justice Ivan Rand and the Classical Legal Tradition

The New Digest is thrilled to feature this guest contribution from Mr Kerry Sun. Mr Sun is a probationary research student at Merton College, the University of Oxford and a Fortescue Scholar with the Canterbury Institute. He has published articles on private and public law theory and jurisprudence.

Recently, an Ontario court made headlines for dismissing the judicial review application of Dr Jordan Peterson against a decision of his professional regulator, the College of Psychologists of Ontario, directing him to undergo mandatory social media training. The College issued the directive based upon concerns that Peterson had made various “degrading, demeaning and unprofessional” comments on Twitter and in public appearances while identifying himself as a clinical psychologist. Key to the Divisional Court’s ruling, on judicial review, was its contention that the College’s directive fell within its statutory mandate to regulate the profession in the “public interest”, and its dismissal of Peterson’s objection that the College was “operating at the very margins of its mandate” or regulatory jurisdiction.

In this essay, I want to consider how the classical legal tradition might bear upon the legal issues underlying this controversial ruling, namely, the jurisdiction of professional regulators to impose restrictions on the public commentary of its members and the grounds for judicial review of such regulation. Critics of the classical tradition have often taken to accusing it of countenancing executive tyranny through (inter alia) an inordinate deference to the administrative state. This impression, while not entirely incomprehensible, is mistaken. To show this I will draw upon a landmark case in Canadian administrative law, a decision which turns out to be consonant with a classical understanding of law and legal form.

In the famous case of Roncarelli v. Duplessis (1959), the plaintiff, a Montreal restaurant proprietor, claimed that the Premier of Quebec had abused his power by ordering the cancellation of his restaurant’s liquor permit. Without recounting the case in its entirety, it suffices to note that the Quebec liquor statute expressly provided that the Liquor Commission “may cancel any permit at its discretion.” Nonetheless, the plaintiff contended that the defendant’s actions were unlawful; he claimed that the Premier acted out of personal animus and for reasons wholly extraneous to the purposes of the regulatory licensing regime, i.e., to punish the plaintiff for supporting the activities of Jehovah’s Witnesses in Montreal.

On appeal, the Supreme Court of Canada gave judgment for the plaintiff and established the legal principle that every legislated grant of decision-making power is inherently bounded, and that “there is no such thing as absolute and untrammelled” administrative discretion. The celebrated, yet enigmatic, opinion of Rand J. in that case bears quoting:

The field of licensed occupations and businesses of this nature is steadily becoming of greater concern to citizens generally. It is a matter of vital importance that a public administration that can refuse to allow a person to enter or continue a calling which, in the absence of regulation, would be free and legitimate, should be conducted with complete impartiality and integrity; and that the grounds for refusing or cancelling a permit should unquestionably be such and such only as are incompatible with the purposes envisaged by the statute … A decision to deny or cancel such a privilege lies within the “discretion” of the Commission; but that means that decision is to be based upon a weighing of considerations pertinent to the object of the administration.

In public regulation of this sort there is no such thing as absolute and untrammelled “discretion” … no legislative Act can, without express language, be taken to contemplate an unlimited arbitrary power exercisable for any purpose, however capricious or irrelevant, regardless of the nature or purpose of the statute. Fraud and cor­ruption in the Commission may not be mentioned in such statutes but they are always implied as exceptions. “Discretion” necessarily implies good faith in discharging public duty; there is always a perspective within which a statute is intended to operate; and any clear departure from its lines or objects is just as objectionable as fraud or corruption. … To pass from this limited scope of action to that of bringing about a step by the Commission beyond the bounds prescribed by the legislature for its exclusive action converted what was done into his personal act.  [emphasis added]

For our purposes, the judgment in Roncarelli v. Duplessis presents a striking affirmation of the classical understanding of legal form, along with its implications for judicial review of administrative action.

Legal Form in Judicial Review

In the classical conception, form is the “constituent principle by virtue of which a thing is determined to be of a specific kind.”[1] It encapsulates a set of characteristics that impart to a given entity its identity and unity, distinguishing it from entities of other kinds. To classify a collection of characteristics under the rubric of form, then, is to render those characteristics intelligible as the components of a unity—that is, as elements that relate to and presuppose each other as parts of a coherent whole (much as, for example, the meat of an apple presupposes the seeds, and vice versa). In short, form is a universal that makes something an instance of this rather than of that. Conjoined with the particular matter or content of an entity which it determines, form completes the essence of the entity.[2]

On the classical view, law is a work of reason, promulgated by an authority who has care of the community, for the common good. This formulation articulates an account of the form of law. It captures those characteristics which give a “law” its specifically legal identity, and it exhibits the intimate interrelationship between law and justice within the classical legal tradition, as legal justice forms part of the common good. As “a specific kind of actualization or performance of human reason,” law is distinguished from the product of pure will.[3] Nor, being oriented to the common good, is it merely an instrument for the advancement of private interests. Thus conceived, the classical notion of legal form is not merely procedural (e.g., concerned with the validity of the enactment), but carries substantive implications for the content of the law.

Now, some have cogently argued that the Roncarelli judgment was founded on postulates about the internal morality of law and values inherent in legality—ideas which continue to motivate contemporary theorising on administrative law.[4] In what follows, I wish to develop these lines of thought further, showing how a classical notion of legal form is latent within this landmark judgment in Canadian law. To do so, we turn to three facets of Rand J.’s reasoning.

First, Rand J. invokes the inherently teleological nature of legal ordinances, insisting that administrative discretion must be tethered to “the nature or purpose of the statute.” In fact, he declares that “there is always a perspective within which a statute is intended to operate.” In other words, this perspective delineates the ambit of an administrator’s discretion. But the idea of such a perspective—one which incorporates substantive requirements of “good faith” and proscribes “fraud” or “corruption”—is senseless if one considers the bare text to be the exclusive source of the law.[5] Even where those constraints are not expressly mentioned (as with the Quebec liquor statute), he maintains that the legislature must have “intended” the statute to incorporate such a perspective and for the discretion to conform to its purposes. In the following sentences, Rand J. goes further and calls it a “distort[ion]” of the “ordinary language of the legislature” to interpret it as conferring an unbounded discretion. But why would it be a “distortion”?

This leads to the second feature of Rand J.’s reasoning, namely that the idea of such a “perspective” signals certain presuppositions about the nature of legislative intent. More precisely, it presupposes that legislation is a reasoned act that aims to secure, in Rand J.’s words, the “interest of the public”—i.e., a modern expression for the common good. It is for this reason that public regulation by law cannot countenance exercises of power that are directed to the “arbitrary,” “capricious or irrelevant” whims of some official. As Rand J. suggests, that would amount to a distortion of legal form; it strips the exercise of official power of its legal essence, “convert[ing] what was done into [the defendant’s] personal act.” Evidently, this line of thought accords with the classical account of law as intrinsically reasoned and oriented to the common good. The underlying presuppositions of legal form make it intelligible to presume, notwithstanding the bare text of the statutory provision, that a legislative enactment cannot normally “be taken to contemplate an unlimited arbitrary power exercisable for any purpose.” What is posited is to be understood in accordance with the precepts of juridical reason.

Third, and most significantly, the judgment brings together the teleological and reasoned dimensions of legal form in its approach to administrative law. At the outset, referring to “[t]he field of licensed occupations and businesses,” Rand J. emphasizes that the administrative powers in question concerned “a calling which, in the absence of regulation, would be free and legitimate.” To my mind, this is a telling remark. It implies the recognition that there is, in fact, some “calling” that exists independently of the regulation or statute. It follows that what characterizes such a “calling”—its distinctive techniques, goods, and virtues—cannot be something merely posited by the lawmaker. Rather, following an Aristotelian school of thought, we might think of a calling, such as that of restaurant proprietor, as a distinctive type of practice that embodies a telos or unity of its own. Given its distinctive character, it stands to reason that each calling, i.e., one of the occupations or businesses among the “field of licensed occupations and businesses,” will involve a distinctive kind of contribution to the common good. There are standards of excellence that are peculiar to the telos of each practice, and the maintenance of such standards is essential to the unity and coherence of that practice and to realizing its distinctive contribution to the common good.

These considerations serve to illuminate a somewhat elusive aspect of the judgment, which relates to the meaning of the “perspective within which a statute is intended to operate.” Although Rand J. says little about the specific content of this “perspective,” we might interpret the judgment, from a classical lens, as embracing a form of professional ethics as the point of regulatory licensing. On this view, it is the telos of the calling that furnishes the relevant meaning of the “perspective”. For it is only from such a perspective that one can identify the standards of excellence, as well as the distinctive contribution to the common good, of the licensed occupation or business as such—i.e., those considerations “pertinent” to one’s fitness to hold a license, and which define the proper ambit of the regulator’s jurisdiction. Thus, in Roncarelli, the defendant stepped outside the perspective of the statute precisely because he revoked the plaintiff’s permit not for any reason related to “his fitness or unfitness to hold a liquor license,” but for his exercise of “an unchallengeable right totally irrelevant to the sale of liquor.”

Seen in this light, Rand J.’s judgment represents an elaboration of the implications of legal form in the specific context of regulatory licensing. Stated simply, the form of these regulatory laws is that of an ordinance oriented to the contribution to the common good embodied in a particular calling, which is a practice with its own distinctive ends. The point of the law is to ensure the integrity of that practice, for the sake of the common good (as Rand J. states elsewhere in his judgment, “for all legitimate purposes of the populace”). It is for this reason that “public regulation of this sort” must eschew the possibility of “absolute and untrammelled discretion.” An unbounded discretion to regulate an occupation qua occupation is unintelligible, for it defies the very notion of ethical standards internal to the regulated occupation. Conversely, where the exercise of official power is directed to these ends, it falls within the perspective of the statute in question and is, in form and substance, a public regulation of a calling.

Thus, Rand J.’s judgment in Roncarelli encapsulates three relevant aspects of legal form: (i) it is teleological, such that the purposes of an enabling statute determines the scope of administrative discretion under it; (ii) it is oriented to the common good or public interest, which excludes administrative action on arbitrary, capricious, or irrelevant grounds; and (iii) in the context of licensed occupations and businesses, it is characterized by a perspective which derives from the classical view of such a calling as an ethical practice that makes a distinctive contribution to the common good, and which informs the scope of the official’s discretion.

The Retreat from Roncarelli

Despite its canonical status in Canadian jurisprudence, the classical understanding of form and the “perspective of the statute” embodied in Roncarelli v. Duplessis has gradually eroded. In recent years, the conception of discretion as animated by a teleological, reasoned, and ordered perspective has been reduced to a thin requirement of “non-arbitrariness”, which is based on the tacit assumption that the scope of regulatory jurisdiction is no more than a product of contingent will. In the former conception, the lawful exercise of administrative discretion exhibits a coherence and unity, rooted in the aim of upholding the telos of the licensed occupation or business. Under the more recent approach, however, administrative power is deemed lawful just so long as it is plausibly referable to some heterogeneous admixture of “public policy” considerations—regardless of whether it conforms to a coherent conception of the mandate of a professional regulator, or exhibits the distinctive form of professional regulation.[6]

Most conspicuously, in Law Society of British Columbia v. Trinity Western University (2018), British Columbia’s legal profession regulator denied licensing accreditation to a Christian university’s proposed law school, on the ground that it had adopted a “Community Covenant” prohibiting sexual intimacy outside of traditional marriage. The Supreme Court of Canada upheld the denial, dismissing the university’s argument that this concern was wholly extraneous to matters of professional competence and fell outside the regulator’s jurisdiction. According to the majority of the Court, “[t]he LSBC’s overarching statutory object in s. 3 of the [Legal Profession Act] — to uphold and protect the public interest in the administration of justice — is stated in the broadest possible terms.” It deemed that “the public interest” in this sense might encompass goals unrelated to the competence of lawyers as such, including “ensuring equal access to the legal profession, supporting diversity within the bar, and preventing harm to LGBTQ law students” and “upholding a positive public perception of the legal profession.”

Yet this account of the “public interest” mandate is profoundly inimical to the idea, enunciated in Roncarelli v. Duplessis, that public administration of a licensed occupation imports a coherent and rationally ordered perspective. As the dissenters, Côté and Brown JJ., noted, the Law Society was constituted not as “a roving, free-floating agent of the state,” but as a professional regulator. Citing Roncarelli, they reasoned that the Law Society’s denial of accreditation was irreconcilable with the perspective appropriate to a professional regulator:

It is readily apparent that the approval of law faculties is tied to the purpose of assessing the fitness of an individual applicant for licensing. … A careful reading of the [Legal Profession Act] reveals that the scope of the LSBC’s mandate is limited to the governance of the practice of law. The LPA’s provisions only relate to matters relevant to the governance of the legal profession and its constituent parts (the LSBC, lawyers, law firms, articled students and applicants). … The LSBC does not enjoy a free-standing power under its “public interest” mandate to seek out conduct which it finds objectionable, howsoever much the “public interest” might thereby be served.

As the dissentients pointed out, the majority’s expansive understanding of the regulator’s “public interest” jurisdiction resulted from a failure to grasp the character of a professional regulator. As such an entity, the Law Society’s mandate is inherently constrained by the perspective of regulating its licensees qua lawyers. By construing the “public interest” mandate of the Law Society widely to encompass virtually any policy objective, the majority could deflect claims that the denial of accreditation was in a strict sense ‘arbitrary’; but in doing so, it effectively inaugurated the very “absolute and untrammelled discretion” that was repudiated in Roncarelli v. Duplessis. Taken seriously, this logic effaces any intelligible distinction between considerations “pertinent” and “irrelevant” to occupational licensing.

Put differently, the underlying dispute in Trinity Western concerned the form of legal regulation: What makes a directive issued by a licensing body identifiable as professional regulation, rather than legislation at large? The classical tradition, of course, is no friend to the usurpation of the lawful authority of those properly charged with the care of the whole community. But for the majority in Trinity Western, any distinction between regulation and legislation is in name only. Even a regulator ostensibly tasked with overseeing a profession can exercise its powers at large to promote the “public interest”, regardless of whether the relevant public interest is affected by acts internal to the profession. In contrast, for the dissenting justices who relied on Roncarelli, legislated grants of discretion are always to be taken as reasoned acts, animated by a perspective which, in the licensing context, is permeable to the telos and internal morality of the licensed occupation. In turn, that perspective necessarily informs the scope of the professional regulator’s jurisdiction, which is addressed to the practitioners of a profession qua professionals.

Now to be sure, the judgment in Roncarelli cannot be taken to authorize courts to invoke free-standing principles of morality to invalidate capacious delegations to the executive. To the contrary, Rand J. limited himself to stating that no legislature should be taken to intend, “without express language,” to embrace an absolute discretion. Yet the point I wish to emphasize, here, is that the key presumption in Roncarelli, the presumption that “there is always a perspective within which a statute is intended to operate,” is sensible only upon presuppositions that dovetail with a classical understanding of legal form. Absent an understanding of law as permeated by reason, there seems to be scant ground for holding, as Rand J. did, that it is a “distortion” of the “ordinary language of the legislature” to interpret it as conferring an unbounded discretion.

Regulatory Jurisdiction and Professional Discipline

In the wake of the Peterson decision, attention has focused on the appropriate “balancing” of free expression rights against a regulator’s statutory mandate to enforce professional standards in the public interest. The Ontario Divisional Court has itself conceived the dispute in this way, opening its judgment by announcing a “clash between a regulated clinical psychologist’s right to speak in a certain manner and the regulator’s power to require the member to moderate that speech.” This framing of the dispute reflects the modern discourse on rights, which conceives a right as an individual entitlement, the scope of which is defined broadly—but in isolation from a conception of rightful relationships between persons in community. As a result, the pursuit of virtually any community interest is considered to “infringe” the right, which must then be artificially curtailed in order to bring it into “balance” with that interest.

From the classical standpoint, such a framing is inapt, in so far as it attributes a false absolutism to the free speech “right” or “value” in question and suggests it is necessarily opposed to the legitimate maintenance of professional discipline. Undeniably, constraints on expressive conduct are part and parcel of the role morality of certain professions. Just as the disclosure of client confidences, for instance, falls outside the scope of lawyers’ expressive rights, health professionals are subject to certain constraints on their expressive freedom by virtue of their role as professionals. In contrast to the modern discourse, the classical view holds that rights are the expressions of justice in community, which are intrinsically conditioned by the imperatives of just and rightful relationships between persons. Properly speaking, the expressive rights of professionals, such as lawyers and psychologists, do not compete with the requirements of professional competence, but account for them in the scope of the right itself. The line between legitimate professional discipline and an abuse of discretion is to be found, not in the impact on expression per se, but rather in the ends for which the discipline is imposed.

Accordingly, the discourse of “balancing” risks eliding the more fundamental and antecedent issue, the ambit of a professional regulator’s jurisdiction. As the preceding discussion of Roncarelli suggests, this implicates two questions: What is the perspective within which the regulator’s enabling statute was intended to operate? And were the regulator’s activities within this perspective? In the Divisional Court’s view, the College was “fulfilling [its] statutory mandate to regulate the practice of psychology in the public interest.” Applying the precedent in Trinity Western, it accepted the College’s position that its concerns about Peterson’s conduct “related to the public interest in members of the College avoiding the use of demeaning or degrading language.” Then, in response to Peterson’s objection that his public commentary lay “at the very margins” of the College’s mandate, the Court noted cursorily that the College’s Code of Ethics “explicitly addresses ‘public statements’” and that it was entitled to regulate public statements whenever individuals identify themselves as clinical psychologists. The unstated implication of this logic is stark—the College is entitled to regulate the public commentary of its members whether or not the content of that commentary relates to the profession itself.

Nonetheless, these responses seem to be rooted in a positivistic assumption that a professional regulation is a formless vessel that can be filled with whatever content has been designated by the relevant authority. The Divisional Court’s judgment discloses no attempt to grapple with the perspective appropriate to the College’s mandate as a professional regulator. Referring to a tweet by Peterson calling a Canadian political advisor a “prik [sic],” for example, the Court endorsed the College’s contention that “the language raised professionalism concerns” due to “the harm from the language used.” Lost in the Court’s reasoning is a grasp of the distinction central to Roncarelli, namely between the standards of excellence peculiar to a licensed field, which are amenable to regulation for legitimate purposes, and those considerations which are irrelevant.

Whatever the specific merits of the Peterson case, the lesson of Roncarelli is that administrative discretion is and must be, in principle, constrained according to an animating perspective. As such, the determining principle of the jurisdiction of professional regulators is located in the underlying connection between the licensing scheme and the telos and integrity of the practice it regulates. Certainly, the boundaries of what falls within the perspective of professional regulation, as opposed to legislation at large, will not always be clear-cut. It may be necessary in many cases for the regulator to engage in the act of determination, by crafting specific rules and regulations that concretize more abstractly-framed standards of professional competence. At times, then, it can be appropriate to defer to the regulator’s own views about the province of professional competence and the attendant scope of its regulatory jurisdiction.

But while some deference may be warranted, it may in some cases be equally appropriate for the legislature to intervene. This is particularly the case where there is a risk that overbearing constraints on professionals, such as disciplinary action over public commentary unrelated to one’s profession, may actually work to the detriment of the common good. Despite the open-endedness of statutory references to the “public interest”, Roncarelli itself demonstrates that administrators can exceed the intended perspective of the licensing regime. In light of these risks, the legislature might adopt amendments to the relevant statutory regime in order to specify the particular grounds on which professionals, like lawyers and psychologists, are subject to professional discipline. A well-crafted legislative specification would not only bring greater coherence to the statutory mandate of professional regulators by clarifying that they are only to regulate members qua professionals, but in doing so, reverse the effects of recent judicial incursions on the essential principles articulated in Roncarelli.

Although its critics deploy the spectre of executive “power grabs”, the classical tradition eludes such hasty caricatures. In taking the law seriously as a work of reason, it embraces a robust conception of legal form that shapes and determines the content and exercise of official power. The irony of such critiques is that the classical notion of form is more stringent than the positivist’s adherence to a procedural notion of form as legal validity. And formalism, above all, is an approach to legal method, not to results. As our examination of Roncarelli v. Duplessis illustrates, it is precisely this attention to form and its juridical implications that serves to delineate rightful from abusive administrative power.

---

[1] David S. Oderberg, Real Essentialism 68 (2007).

[2] The description offered in this and the following paragraph draws closely upon the discussion of form in Ernest J. Weinrib, Legal Formalism: On the Immanent Rationality of Law, 97 Yale L.J. 949, 958–61 (1988); David S. Oderberg, Real Essentialism 65–71 (2007); Brian M. McCall, The Architecture of Law: Rebuilding Law in the Classical Tradition 375–412 (2018). See also Konstanze von Schütz, Immanent Ratio Legis? Legal Forms and Statutory Interpretation, in Ratio Legis: Philosophical and Theoretical Perspectives 161 (Verena Klappstein & Maciej Dybowski eds., 2018).

[3] Martin Rhonheimer, Natural Law as a “Work of Reason”: Understanding the Metaphysics of Participated Theonomy, 55 Am. J. Juris. 41, 42 (2010).

[4] See, e.g., David Dyzenhaus, The Deep Structure of Roncarelli v. Duplessis, 53 U.N.B.L.J. 111 (2004); Cass R. Sunstein & Adrian Vermeule, Law and Leviathan: Redeeming the Administrative State (2020).

[5] See Dyzenhaus, supra, at 131–32. In this regard, the reasoning of Rand J. is thrown into sharp relief by the dissenting opinion of Cartwright J., who claimed he was “unable to find that the Legislature has, either expressly or by necessary implication, laid down any rules to guide the commission as to the circumstances under which it may refuse to grant a permit or may cancel a permit already granted.”

[6] I am grateful to Stéphane Sérafin for impressing upon me the significance of this point. In many ways, the denaturation of the rational and ordered conception of the “perspective of the statute” in Roncarelli v. Duplessis into a reductive notion of “non-arbitrariness”, reliant upon a heterogeneous and disarticulated collection of policy rationales, is reminiscent of MacIntyre’s account of the abandonment of teleological context in moral theory and its consequences. See Alasdair MacIntyre, After Virtue: A Study in Moral Theory 51–61 (3d ed. 2007).