Human Rights in Administrative Decision-Making I: Discretion and Judgement in Statutory Interpretation

Last month I gave a talk at the Journal of Commonwealth Law’s symposium on Human Rights in Administrative Agencies at the Université de Montréal, entitled “The Inevitability of Administrative Discretion and What it Should Mean for Human Rights”. Here’s the abstract:

Given the vagaries of language and life there is no such thing as “clear” legislative text. Empirical research also indicates that hierarchical superiors in administrative agencies are pathologically incapable of directing subordinates. As a result, administrative discretion is inevitable. The question then becomes how front-line decision-makers should exercise this discretion. Should they ignore human rights issues? Should they address these issues as lawyers would, even if they have no legal training? I argue that the answer to both of these questions must be no. Moreover, I note that the Charter values framework from Doré has a solid basis in the on-the-ground realities of administrative discretion. Furthermore, I insist that the ability of non-lawyers working in administrative agencies to give meaningful content to constitutional commitments is a valuable contribution to democracy, broadly understood. This is not to say that courts should necessarily defer to administrative decision-makers on constitutional questions, which is entirely a separate matter, but to say that administrative discretion should presumptively be exercised in a large and liberal way, not characterized by pedantry and pettifoggery.

And here’s a draft of Part I (with Parts II, III and IV to follow). Comments welcome!

Legislation is sometimes vague or ambiguous. Lawyers tend to describe any degree of indeterminacy as “ambiguity”, but this is incorrect: “Ambiguity in the strict sense – with a stark choice between given interpretations – is comparatively rare in legislative drafting”.[1] Typically, lawyers struggling to give meaning to a statutory provisions have been confronted by vagueness: “The difficulty with vague concepts is in deciding whether some apparent instance at the margins is inside or outside of the concept”.[2] Especially where an administrative decision-maker has been tasked with implementing a statutory scheme, “legislation is often deliberately worded vaguely, the intention being that vaguely worded standards will be made more concrete by the decisions of specialized administrative actors, who are best placed to develop the relevant policies”.[3] Old-style ‘command-and-control’ regulation has been replaced by new governance techniques. Already in the 1990s, regulatory theorists could recognize that discretion was being distributed “within rules in which legal actors may exercise choice” through the “conscious and strategic use of terms which are vague as to manner, time or place, are evaluative or use generic terms to act as implicit derogations…”[4] The process has only accelerated since then.

Where legislation is ambiguous or (more commonly) vague, it is inevitable that an administrative decision-maker will exercise discretion or judgement in putting the legislation into effect. As Lord Hoffmann observed in Moyna v Secretary of State for Work and Pensions,“[i]n any case in which a tribunal has to apply a standard with a greater or lesser degree of imprecision and to take a number of factors into account, there are bound to be cases in which it will be impossible for a reviewing court to say that the tribunal must have erred in law in deciding the case either way”.[5]

Even where a statute is not vague or ambiguous, discretion or judgement on the part of interpreters remains unavoidable, as all language is open-textured. As HLA Hart put it, “[i]n all fields of experience, not only that of rules, there is a limit, inherent in the nature of language, to the guidance which general language can provide”.[6] The philosopher JL Austin brilliantly captured the open texture of language with his example of a “goldfinch” which explodes or begins to quote Virginia Woolf.[7] The point is that a “goldfinch”has obvious standard instances but the term becomes indeterminate once we are faced with a goldfinch that has exploded: “It is the possibility of vagueness – the potential vagueness – of even those terms that appear to have no uncertainties with respect to known or imagined applications”.[8] Our existing language conventions are no longer capable of describing reality.

When it comes to legal language, lawyers are less concerned with exploding goldfinches than with scenarios in which giving effect to the plain meaning of a statutory provision would lead to a problematic result: “In such cases we do not have linguistic open texture, but we might have legal open texture, and it is in such cases that legal decisionmakers must decide what to do”.[9] In legal systems the potential for the text and purpose of a provision to come into conflict is “ever present”: “If the straightforward reading of the law produces a ridiculous or even merely suboptimal outcome, are legal actors required or even permitted to reach the right outcome instead of the outcome seemingly mandated by the plain meaning of the words on the page?”[10] The famous debate between Professor Hart and Professor Lon Fuller about the nature of law turned in part on the ‘no vehicles in the park’ hypothetical:

A legal rule forbids you to take a vehicle into the public park. Plainly this forbids an automobile, but what about bicycles, roller skates, toy automobiles? What about air-planes? Are these, as we say, to be called “vehicles” for the purpose of the rule or not? If we are to communicate with each other at all, and if, as in the most elementary form of law, we are to ex-press our intentions that a certain type of behavior be regulated by rules, then the general words we use – like “vehicle” in the case I consider – must have some standard instance in which no doubts are felt about its application. There must be a core of settled meaning, but there will be, as well, a penumbra of debatable cases in which words are neither obviously applicable nor obviously ruled out. These cases will each have some features in com-mon with the standard case; they will lack others or be accompanied by features not present in the standard case. Human invention and natural processes continually throw up such variants on the familiar, and if we are to say that these ranges of facts do or do not fall under existing rules, then the classifier must make a decision which is not dictated to him, for the facts and phenomena to which we fit our words and apply our rules are as it were dumb.[11]

For Hart, in penumbral cases – bicycles, roller skates or toy automobiles – the decision-maker must “exercise a creative choice between alternatives”,[12] drawing on “various aims and policies” (which are not “law” in the same way as the “core of legal rules whose meaning is settled”).[13] In response to the Fullerian view that, in such instances, judges must look to the true purpose of the rule in question such that penumbral cases involve a natural elaboration of the scope of an existing rule of law, Hart thought “it cannot be doubted that, for most cases of interpretation, the language of choice between alternatives, ‘judicial legislation’ or even ‘fiat’ (though not arbitrary fiat), better conveys the realities of the situation”.[14] For Fuller, by contrast, even where individual words are in issue, the same process – construing words against the purpose sought to be achieved – is operative: “If in some cases we seem to be able to apply the rule without asking what its purpose is, this is not because we can treat a directive arrangement as if it had no purpose. It is rather because, for example, whether the rule be intended to preserve quiet in the park, or to save carefree strollers from injury, we know, ‘without thinking,’ that a noisy automobile must be excluded”.[15] Both protagonists can be understood to accept that discretion and judgment will often be required of interpreters.[16]

Legal open texture can also be observed in scenarios where statutes are interpreted in accordance with constitutional principles. In Riggs v Palmer, the majority of the New York Court of Appeals famously gave overriding effect to the principle than no man shall profit from his own wrong notwithstanding statutory language which clearly permitted a beneficiary of a will who murdered the testator (in order to prevent the latter from changing the will).[17] In two well-known apex-court decisions from the United Kingdom – Anisminic v Foreign Compensation Commission[18] and R (Privacy International) v Investigatory Powers Tribunal[19]– a majority of judges refused to give effect to a statutory provision ousting judicial review of executive action, decisions which have been understood to have been grounded in the rule of law and/or separation of powers. In these cases, statutory language which was clear as a linguistic matter – the murderer can inherit; the courts cannot consider the lawfulness of certain administrative decisions – became open textured when the constitutional implications of clarity were considered. The interpreters had to exercise discretion and judgement in ascertaining the scope of the statutory provisions at issue.

Open texture – whether of the linguistic or legal variety – can be observed in other areas too. Background context may render unclear language which seems, at first glance, pellucidly clear.[20] On highways in Montreal, there are prominent signs proclaiming “MAX: 100; MIN: 60”. The language here is clear: one may not exceed 100km/h and one must exceed 60km/h. What if, however, you are stuck in traffic? When I lived in Montreal, I was often stuck in traffic and often had plenty of time to stare at those signs and ponder their meaning. As a linguistic matter, when snarled up in traffic I was violating the rule. Obviously, though, no traffic cop would give me a ticket and, in the unlikely event that one did, no judge would uphold it. This even though the rule was linguistically clear. Those charged with enforcing the rule therefore must exercise discretion and judgement.

Sometimes, unforeseen scenarios can create open texture. Here I have in mind not merely the unfortunate applications of a clear rule – as in Riggs v Palmer – but truly outlandish scenarios like the one envisaged by Anthony D’Amato in “Aspects of Deconstruction: The ‘Easy Case’ of the Under-Aged President”.[21] It seems clear that someone under the age of 35 cannot become president of the United States[22] until “we are willing to supply a suitable context in which we can imagine that an eighteen year-old would be nominated and have a serious chance of winning the election”.[23] A “contextual change” – a student-led revolt or a disease which wipes out everyone over the age of twenty – permits the making of innovative and unforeseeable arguments, which would previously have been classified as “outrageous”.[24] Whether we describe this as an instance of linguistic or legal open texture,[25] the point is that even when we accept that language is clear, in some instances those charged with interpreting it will have to exercise discretion and judgement (in giving effect to the formerly outrageous argument or sticking fast to the clear meaning of the provision at issue).[26]

Experience suggests, moreover, that the meaning of statutory language is often not “clear” at all, certainly not when courts are called upon, for where there is litigation it usually follows that, “the words are ambiguous enough to induce two people to spend good money in backing two opposing views as to their meaning”.[27] It has even been doubted whether, as an analytical matter, statutes can ever be “clear”. As Max Radin argued: “As a matter of fact, in most cases when courts say that a statute is plain and therefore needs no interpretation, they do so in the inverted fashion which marks so much of the judicial process. They have already interpreted, and they then declare that so interpreted the statute needs no further interpretation”.[28] The point is not that law is in some way radically indeterminate but that significant discretion and judgement is often exercised by administrative officials, even in situations where the statutory language at issue might be said to be “clear”.

[1] Pong Marketing and Promotions Inc. v. Ontario Media Development Corporation, 2018 ONCA 555, at para. 46, per Miller JA.

[2] Pong Marketing and Promotions Inc. v. Ontario Media Development Corporation, 2018 ONCA 555, at para. 47, per Miller JA.

[3] Pong Marketing and Promotions Inc. v. Ontario Media Development Corporation, 2018 ONCA 555, at para. 48, per Miller JA.

[4] Julia Black, Rules and Regulators (Oxford: Clarendon Press, 1997), at p. 216.

[5] Moyna v Secretary of State for Work and Pensions [2003] 1 WLR 1929, at para. 20.

[6] H.L.A. Hart, The Concept of Law, 2nd ed. (Oxford: Oxford University Press, 1994) at p. 126.

[7] J.L. Austin, “Other Minds” (1946) 20 Proceedings of the Aristotelian Society 148, at p. 160.

[8] Frederick Schauer, “On the Open Texture of Law” (2013) 87 Grazer Philosophische Studien 197, at p. 199.

[9] Frederick Schauer, “A Critical Guide to Vehicles in the Park” (2008) 83 NYU Law Review 1109, at p. 1128.

[10] Frederick Schauer, “A Critical Guide to Vehicles in the Park” (2008) 83 NYU Law Review 1109, at p. 1116.

[11] HLA Hart, “Positivism and the Separation of Law and Morals” (1958) 71 Harvard Law Review 593, at p. 607.

[12] HLA Hart, “Positivism and the Separation of Law and Morals” (1958) 71 Harvard Law Review 593, at p. 612.

[13] HLA Hart, “Positivism and the Separation of Law and Morals” (1958) 71 Harvard Law Review 593, at p. 614.

[14] HLA Hart, “Positivism and the Separation of Law and Morals” (1958) 71 Harvard Law Review 593, at p. 629.

[15] Lon Fuller, “Positivism and Fidelity to Law: A Reply to Professor Hart” (1958) 71 Harvard Law Review 630, at p. 663.

[16] See also Julia Black, Rules and Regulators (Oxford: Clarendon, 1997), chapter 1, on the indeterminacy of rules.

[17] (1889) 22 NE 188 (NY).

[18] [1969] 2 AC 147.

[19] [2019] UKSC 22, [2019] 2 WLR 1219.

[20] See also Mannai  Investment  Co  Ltd v. Eagle  Star  Life Assurance  Co  Ltd [1997] AC 749, where the issue was whether a tenant who gave notice to terminate a lease on “January 12” should have been understood to give his notice on January 13 (consistent with the terms of the lease). A majority of the House of Lords held, having regard to the background context, that “January 12” meant “January 13”. As Lord Hoffmann subsequently explained in Investors Compensation Scheme Ltd v. West Bromwich Building Society, [1998] 1 WLR 896, at pp. 912-913, background context “may not merely enable the reasonable man to choose between the possible  meanings  of  words  which  are  ambiguous  but  even  (as  occasionally  happens  in  ordinary  life)  to conclude  that  the  parties  must,  for  whatever  reason,  have  used the  wrong  words  or  syntax”.

[21] (1990) 84 Northwestern University Law Review 250.

[22] Article II, s. 1, clause 5: “No person…shall be eligible to the Office of President…who shall not have attained to the Age of thirty-five Years…”

[23] D’Amato at p. 254.

[24] D’Amato at p. 255.

[25] For Frederick Schauer, “On the Open Texture of Law” (2013) 87 Grazer Philosophische Studien 197, at p. 202 the difference between Austin’s exploding goldfinch and Hart’s toy automobile “is that we can not only imagine the latter, but can also describe it with existing linguistic tools”.

[26] Frederick Schauer, “On the Open Texture of Law” (2013) 87 Grazer Philosophische Studien 197, at p. 212 observes: “Defeasibility is…not a property of rules at all, but rather a characteristic of how some decision-making system will choose to treat its rules” (emphasis original). Professor Schauer argues that interpretive discretion could be removed by having recourse to closure rules or (more simply) by insisting that interpreters give effect to the clear language of statutory provisions. In the present context I would observe, first, that adding closure rules to existing linguistic conventions is unlikely to eliminate interpretive discretion: as discussed in Part II, the more rules there are, the more scope there is for front-line decision-makers to exercise discretion and judgement. Second, the argument for giving effect to clear language is itself based on constitutional principles (accord Mark Greenberg, “Beyond Textualism” UCLA School of Law, Public Law Research Paper No. 19-41). As such, the Schauerian desire to give effect to linguistic meaning would have to be balanced against competing constitutional principles – like, as in Anisminic and Privacy International, the rule of law – in a scenario pitting the clear language of a rule against its constitutionally problematic consequences. Moreover, as Julia Black argues, Rules and Regulators (Oxford: Clarendon, 1997), at p. 217:

Increasing precision is an attempt to render explicit the conduct expected in as wide a range of circumstances as possible. It is an attempt to substitute rules for the tacit understandings and informed reading which rules need, but which may not exist. As such, it can only fail: increased precision may reduce but can never eliminate the inherent indeterminacy of rules and does not in the end create the understanding which it is trying to replace.

[27] John Willis, “Statute Interpretation in a Nutshell” (1938), 16 Can. Bar Rev. 1, at pp. 4-5.

[28] Max Radin, “Statutory Interpretation” (1930) 43 Harvard Law Review 863, at p. 869. See similarly Ronald Dworkin, Law’s Empire (Harvard University Press, Cambridge MA, 1986), at p. 353.

This content has been updated on November 22, 2019 at 03:36.