A Defence of Administrative Law Doctrine, Part II.B, Legality, Rationality and Procedural Propriety

This is an extract from my paper “A Defence of Administrative Law Doctrine

The second proposition is that the heads of review – legality, rationality and procedural propriety – are too abstract to guide judges: “Though a convenient summary of our conclusions in any particular case, the doctrine cannot provide any justifying ground for those conclusions”,[1] as “their application to the facts of any particular case–and hence their concrete meaning–is largely dependent on the specific context under review”.[2] They are not, in other words, capable of precluding particular outcomes in individual cases.

There is significant force to this proposition. I propose to take the heads of review in the order in which the proposition has greatest force: rationality, legality and then procedural propriety. In respect of each head of review, however, I will suggest that Professor Allan’s proposition can only be partially accepted.

(i)                Rationality

In Wednesbury unreasonableness, Professor Allan saw a “flexible and contextual…standard of review” which permits “varying levels of judicial scrutiny or substitution of judgment, according to context”.[3] Moreover, any difference between Wednesbury unreasonableness and proportionality “depends on the degree to which the court is prepared to make its own assessment of the balance of interests in judging rationality”, such that there is no difference between the two in concrete terms.[4]

There is no doubt that the various formulations of irrationality or unreasonableness which peppered the 20th century law reports are colourful but largely lacking in analytical content. They may be “salutary reminders of the appropriate limits of judicial oversight”[5] but give little guidance in concrete cases:

  • “if a decision on a competent matter is so unreasonable that no reasonable authority could ever have come to it”.[6]
  • “whether the impugned decision plainly and unambiguously flies in the face of plain reason and common sense”.[7]
  • “a decision which is so outrageous in its defiance of logic or of accepted moral standards that no sensible person who had applied his mind to the question to be decided could have arrived at”.[8]
  • “so flawed that no amount of curial deference can justify letting it stand”.[9]

Professor Allan is hardly the only scholar to have besmirched judicial articulations of rationality as a head or ground of review. The language of emptiness and pernicious would be apt as applied to this list of formulations, which obscure more than they reveal and place “insurmountable hurdles” between those subject to shoddy treatment and meaningful remedies: “it is very difficult, if not impossible, to come up with real-world cases in which such extreme irrationality has been present”.[10] It is not enough to describe imagined instances of unreasonableness; it is necessary to provide criteria for determining why a particular act was unreasonable.[11]

Contemporary formulations which rely on the concept of “intensity” or the metaphor of a “range” of permissible, acceptable outcomes are little better in this regard than their 20th century forebears: they invite contextual analysis without necessarily constraining it.[12] One must look to the pages of the law journals rather than the law reviews for attempts to provide systematic accounts of the constraints inherent to rationality in its administrative law sense,[13] or for arguments in favour of displacing rationality with a more structured proportionality test.[14]

The Supreme Court of Canada’s detailed articulation of a reasonableness standard in the Vavilov case is a notable exception. The majority judges set out two different types of unreasonableness: a want of reasoned explanation on the one hand; a failure to respect applicable legal and factual constraints on the other.[15] Some aspects of unreasonableness, Canada-style, might have to yield to an Allanite charge of emptiness: an administrative decision-maker’s compliance with the governing statutory scheme and consideration of the important interests of the individual concern are undoubtedly highly contextual.[16] Nonetheless, this detailed articulation of unreasonableness certainly guides and likely constrains reviewing courts. For example, an interpretation of a statute will be unreasonable if it is inconsistent with statutory text, context and purpose;[17] intervention will only be appropriate on factual matters where a decision-maker has “fundamentally misapprehended” the evidence;[18] and an administrative decision-maker’s departure from past practice will only be unreasonable where it lacks justification.[19] These tests for intervention constrain judges, precluding them from finding decisions to be reasonable or unreasonable unless the tests set out in Vavilov are satisfied.

Moreover, the majority decision in Vavilov does provide clear markers in relation to important issues: for one thing, the onus is on the applicant to demonstrate the unreasonableness of a decision;[20] for another, a reviewing court is forbidden from engaging in a de novo assessment of the underlying issue,[21] even on questions of statutory interpretation.[22]

Furthermore, the constraints set out in Vavilov have quickly become encrusted with jurisprudence. Judges in the Federal Court of Canada, for example, are constrained by virtue of horizontal stare decisis to apply their colleagues’ understanding of the requirements of Vavilov as applied to the various subject-matter areas in which they hear judicial review applications, such as immigration, labour relations and environmental law. In concrete terms, the unreasonableness of a decision on immigration law (say) will turn as much if not more on Federal Court jurisprudence than on the majority reasons in Vavilov.

To the extent this can be said of other jurisdictions – though a collective focus on apex court decision-making may obscure the truth – then the charge of emptiness must be qualified as far as rationality is concerned.

(ii)              Legality

As far as “legality” is concerned, Professor Allan has written, “[t]he normative constraints that govern the exercise of an administrative discretion delimit an authority’s jurisdiction as fully as the descriptive criteria that specify its functions; the relevant jurisdiction, properly understood, is simply that field of choice, if any, that remains when all applicable legal limits have been identified”.[23] Very simply, “An error can be properly labelled or classified only in the light of the policies and purposes that inform the statute, construed as an intelligible whole, and the relevant consequences for the case in hand”.[24]

The force of this point has been demonstrated at considerable length, by reference to the caseload of the Administrative Court of England and Wales, by Dr Sarah Nason.[25] Her review of first-instance judicial review decisions highlights that judges rarely engage in detailed discussions of the doctrinal niceties which one finds in the textbooks or lectures on administrative law. Rather, first-instance decisions tend to turn on judicial judgement, informed by the judge’s experience and the arguments of the parties, with the principles of statutory interpretation to the fore.

These findings about “legality” are not especially surprising. In English administrative law, questions of law are invariably reviewed on a correctness standard, with no deference to administrative decision-makers. Questions of law, in turn, require the application by the courts of the principles of statutory interpretation which, in their modern guise, involve an appreciation of text, purpose and context. Any exercise of divining statutory meaning undertaken by a contemporary court will involve careful consideration of a range of factors. The point can be extended to the doctrines of relevancy and propriety. Determining relevant considerations and proper purposes is a matter of statutory interpretation, to be addressed by reference to text, purpose and context. The object of the statutory interpretation exercise “is to ascertain the statutory purpose or object which the draftsman had in mind”.[26]

However, it would go too far to state that doctrine is irrelevant in such cases. The general doctrine states that questions of law are for the courts[27] and, accordingly, requires that the courts must give authoritative answers to legal questions. Even in Canada – a jurisdiction which embraces deference on questions of law, at least some of the time – a correctness standard calls for judicial application of the principles of statutory interpretation unencumbered by any requirement to defer to the analysis of an administrative decision-maker. Doctrine is unlikely to be debated in “legality” cases, but absence of evidence is not evidence of absence: rather, the doctrine sets the framework in which matters of “legality” fall to be assessed.

Moreover, in relevancy cases where an administrative decision-maker has a discretion as to whether to take a particular consideration into account, doctrine dictates how a court is to review the exercise of the discretion,[28] and in both relevancy and propriety cases there are judicially developed doctrines about the consequences of an administrative decision-maker’s reliance on an irrelevant consideration or improper purpose.[29]

In all events, these “legality” doctrines act as constraints on what judges can do in judicial review cases, by ruling in or ruling out certain conclusions: judges cannot (where a correctness standard is appropriate) defer to an administrative decision-maker’s reasonable view of the law; judges cannot add or subtract a relevant consideration unless the statute, properly interpreted, rules the relevant consideration in or out; and judges are not at liberty to determine the consequences of taking into account an irrelevant factor or relying on an improper purpose.

I acknowledge that the phrase “properly interpreted” is doing a significant amount of work in the preceding paragraph. In many cases, the “proper” interpretation of a statutory provision will, on Professor Allan’s view, involve a wide-ranging consideration of context and constitutional principle. But the fact that context and constitutional principle should be (and generally are) taken into account as part of the statutory interpretation exercise does not mean that doctrine is irrelevant in “legality” cases. Doctrine is, rather, inherent to the exercise.

(iii)           Procedural Propriety

The contemporary, context-sensitive approach to procedural fairness has long since displaced the former rules of natural justice. These old rules – nemo iudex in sua cause and audi alteram partem – were too inflexible for the burgeoning administrative state and were cast aside in the latter half of the twentieth century in favour of a duty of “fairness”.

The move from natural justice to fairness has long been a source of angst for administrative lawyers. Professor Loughlin proclaimed a “crisis” in administrative law theory in the late 1970s[30] and more recently Lord Justice Beatson has agonized openly about how a flexible approach to fairness might create a Tennysonian wildnerness of single instances.[31]

Doctrine might, accordingly, seem to be conspicuous by its absence in this area. Indeed, high-level references to fairness and a multiplicity of factors will seldom guide the resolution of concrete cases. Yet it does not follow that doctrine does not matter. Consider the recent work of Dr Joanna Bell.[32] She describes the three features of administrative law which lend it an inherent complexity: administrative law cases involve the interpretation of discrete statutory provisions, not the articulation of sweeping general principles; in administrative law cases, judges juggle a plurality of principles, purposes or values and cannot rely on one unifying meta-concept; and the protagonists in administrative law cases find themselves in different relationships, sometimes tied together by general norms, sometimes by individualized decisions highly specific to the circumstances of the parties. This is the anatomy of administrative law and, she argues, we must appreciate it in order to understand the subject. She deconstructs the heads of judicial review in a manner which Professor Allan would doubtless appreciate.

But Dr Bell’s is not simply an exercise in deconstruction. Recognizing the variability of administrative law does not lead us to a wilderness of single instances. Rather, focusing on first-instance decisions on procedural fairness, legitimate expectation and standing, Dr Bell identifies a significant degree of coherence. Notably, procedural fairness is tied together by heavy reliance on procedural codes and the rights to receive notice and make representations: there is a “series of patterns” in the jurisprudence.[33]

My observation of judicial decisions on procedural fairness matters leads me to similar conclusions: judges are not and do not consider themselves to be at liberty to make ‘all things considered’ judgments about fairness based on the circumstances of individual cases: questions about compliance with the duty of fairness are “not to be asked and answered simply by reference to context: more particular considerations attach to individual procedural fairness rights”.[34] No sensible litigant will go to court and argue that a procedure was “unfair”. Rather, they will argue that a particular procedural right was not accorded, for example the right to notice. But the right to notice (say) comes encrusted in particular contexts with past practice by the administrative decision-maker, policy guidance issued by the decision-maker or others and judicial decisions defining the scope of the right in question. Here, judges are not making decisions based only on context: they are constrained by soft law instruments, past administrative practice and binding judicial precedent. This is a doctrinal superstructure within which judgments about procedural fairness fall to be made.

One might make an analogy here to Professor Allan’s discussion of constitutional conventions. He writes that “interpretation involves judgement”, with the decision-maker required to “take sides between the competing theories or viewpoints”.[35] In respect of rationality, legality and procedural propriety, the judge ‘takes the winning side’ by providing a reasoned justification for his or her conclusions, grounded in the considerations revealed as relevant by doctrine. I suggest, accordingly, that the ‘heads’ of review are not empty but rather constrain judges, sometimes significantly.

[1] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 430.

[2] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 431. See also D.G.T. Williams, “Statute Law and Administrative Law” [1984] Statute Law Review 157, at p. 160:

[Administrative law] cases range over many thousands of statutory provisions: this militates against any effective doctrine of stare decisis and explains why administrative law is primarily a subject of obiter dicta, with hallowed passages, such as the Wednesbury rule or the Carltona principle, offering judges convenient shorthand principles in the application of the ultra vires rule.

[3] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 437. See also The Sovereignty of Law: Freedom, Constitution and Common Law (Oxford University Press, Oxford, 2013), at p. 113: “Wednesbury unreasonableness, in other words, expresses the conclusion of legal analysis, which encompasses all the relevant rule-of-law criteria as they apply to the facts or circumstances in view”.

[4] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 438.

[5] Paul Daly, “The Language of Administrative Law” (2016) 94 Canadian Bar Review 519, at p. 530.

[6] Associated Provincial Picture Houses v Wednesbury Corporation [1948] 1 KB 223, at p. 230.

[7] State (Keegan) v Stardust Victims’ Compensation Tribunal [1986] IR 642, at p. 658.

[8] Council of Civil Service Unions v Minister for Civil Service [1985] AC 374, at p. 410.

[9] Law Society of New Brunswick v Ryan [2003] 1 SCR 247, at para. 52.

[10] Paul Craig, “Proportionality, Rationality and Review” [2010] NZLR 265, at p. 274.

[11] Anthony Lester and Jeffrey Jowell, ‘Beyond Wednesbury: Substantive Principles of Administrative Law’ [1987] Public Law 368, at p. 371.

[12] See also Sarah Nason, Reconstructing Judicial Review (Hart, Oxford, 2016).

[13] See e.g. Rebecca Williams, “Structuring Substantive Review” [2017] Public Law 99; Hasan Dindjer, “What Makes an Administrative Decision Unreasonable?” (2021) 84 Modern Law Review 265.

[14] See e.g. Paul Craig, “Proportionality, Rationality and Review” [2010] NZLR 265. Though see also the dissents in Law Society of British Columbia v. Trinity Western University, 2018 SCC 32, [2018] 2 SCR 293, arguing for the return of proportionality assessments in cases involving alleged infringements by administrative decision-makers of fundamental rights, especially because the proportionality test places an onus on the decision-maker to justify any infringement.

[15] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at paras. 105-138.

[16] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at paras. 108-110, 133-135.

[17] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 120. Cf the comments at paras. 119, 122 and 123, which permit departures from the judicial standard of statutory interpretation.

[18] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 126.

[19] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 131.

[20] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 75.

[21] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 83.

[22] Canada (Minister of Citizenship and Immigration) v Vavilov 2019 SCC 65, [2019] 4 SCR 653, at para. 124.

[23] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 441.

[24] “Doctrine and theory in administrative law: an elusive quest for the limits of jurisdiction” [2003] Public Law 429, at p. 442.

[25] Reconstructing Judicial Review (Hart, Oxford, 2016).

[26] R v Secretary of State for the Environment, Transport and the Regions, Ex p Spath Holme Ltd [2001] 2 AC 349, at p. 381, per Lord Bingham. See e.g. Tesco Stores v Secretary of State for the Environment [1995] 1 WLR 759; R (Palestine Solidarity Campaign Ltd) v Secretary of State for Housing, Communities and Local Government [2020] UKSC 16, [2020] 1 WLR 1774.

[27] R v Hull University Visitor, ex parte Page [1993] AC 682.

[28] Lord Irvine of Lairg, “Judges and decision makers: the theory and practice of Wednesbury review” [1996] Public Law 59, at pp. 67-69. See e.g. R (Rogers) v Swindon NHS Primary Care Trust [2006] EWCA Civ392,[2006] 1 WLR 2649.

[29] Paul Daly, Understanding Administrative Law in the Common Law World (Oxford University Press, Oxford, 2021), at pp. 141-143.

[30] “Procedural Fairness: A Study of the Crisis in Administrative Law Theory” (1978), 28 University of Toronto Law Journal 215.

[31] R (L) v West London Mental Health NHS Trust [2014] EWCA Civ 47, at para. 72.

[32] The Anatomy of Administrative Law (Hart, Oxford, 2020).

[33] The Anatomy of Administrative Law (Hart, Oxford, 2020), at p. 127.

[34] Paul Daly, Understanding Administrative Law in the Common Law World (Oxford University Press, Oxford, 2021), p. 81.

[35] The Sovereignty of Law: Freedom, Constitution and Common Law (Oxford University Press, Oxford, 2013), at p. 55.

This content has been updated on October 7, 2022 at 15:31.