Despotism, Revisited

Thoughts upon belatedly reading an (anti-)administrative law classic

I have, rather belatedly, read an (anti-)administrative law classic, The New Despotism by Lord Hewart’s  ― an attack on the power of what would come to be called the administrative state published in 1929 by the then-Lord Chief Justice of England. The book made quite an impression when it was published, prompting the government to set up an inquiry, and even has its own Wikipedia page. However, I don’t think The New Despotism is often discussed in Canada these days. (A quick HeinOnline search shows no more than occasional citations in the past decade; and, what little that’s worth, I hadn’t heard about it until I sat in on my colleague Vernon Rive’s administrative law lectures.) So perhaps some comments here may be of interest, if only to my fellow dabblers, despite the book’s antiquity.

In a nutshell, Lord Hewart was alarmed by the expansion of unreviewable legislative and adjudicative powers delegated by Parliament to officials within the executive branch. While he is almost certainly skeptical of the administrative state generally, Lord Hewart mostly suspends this skepticism and focuses his attacks not on the exercise of power by administrative decision-makers as such, but on the fact that, all too often, administrative power is exercised more or less secretly, without the persons affected by it being able to make submissions to decision-makers, or without decision-makers having to take these submissions into account, or to explain how they reached the conclusions they did. He criticizes legislation empowering administrators to override statutes, or to interpret and apply them without any judicial oversight. Such legislation, he insists, creates a system that is not, properly speaking, one of “administrative law”, such as it exists in Europe (Lord Hewart doesn’t share A.V. Dicey’s notorious disdain for continental administrative law), but one of “administrative lawlessness”.

The remarkable thing is that, while it is fashionable to describe The New Despotism (insofar as it is referred to at all) as a “tirade” delivered by an apologist for the nightwatchman-state dark ages, his critique has been largely accepted ― including by the latter-day defenders of the administrative state ― and incorporated into modern administrative law. Whatever our views on the Canadian (and American) practice of deference to administrative interpretations of statutes, even those who defend this practice accept that some judicial oversight over administrative decision-makers is constitutionally essential. And they, like their critics, would share Lord Hewart’s indignation at decision-making processes in which anonymous officials may act without receiving evidence or submissions from affected parties, whom they need not appraise of their concerns, and are not required to give reasons. He might not be kindly remembered, but in a very real sense, Lord Hewart won the battle of ideas. Pro- or anti-administrativists, we largely agree with him, and indeed among ourselves. The outstanding disagreements are of course significant, but not nearly as significant as the general assent to the subjection of administrative decision-making to judicial review in matters both procedural and substantive.

Interestingly, however, this consensus was not implemented in the manner Lord Hewart envisioned. It is largely reflected in the development of the common law, and not so much in changes to legislative practice which he urged. Some legislative changes have occurred. In particular, there are better, though I suspect still deficient, mechanisms for Parliamentary review of regulations, which Lord Hewart called for. But legislatures have not ceased purporting to delegate vast and unreviewable powers to the executive. What has changed is that the courts came to take a much more skeptical approach to such legislation, and seldom give it its full effect. This, I think, is not surprising. Lord Hewart thought that, to eradicate administrative lawlessness, “what is necessary is simply
a particular state of public opinion”, for which to “be brought into existence what is necessary is simply a knowledge of the facts”. (148) This seems almost touchingly naïve ― almost, because, as a former politician himself, Lord Hewart ought to have known better. It is implausible that public opinion can be drawn to, let alone firmly focused on, issues that are bound to strike non-lawyers as purely technical matters. This is something worth pondering as we reflect on the relative legitimacy of judicially-articulated and legislated rules, whether generally or specifically in the context of administrative law.

Let me now go back to the disagreement between those who favour judicial deference to administrative decision-makers and those who resist it. That Lord Hewart would surely have been in the latter camp will not persuade anyone who is not, given his reputation as an arch-anti-administrativist. But there is another jurist, whose name carries more authority in Canada than Lord Hewart’s, whom I am happy to claim for non-deferential camp (to which I belong): none other than Lord Sankey, of the “living tree” fame. In an extra-judicial speech, delivered just months before the opinion in Edwards v Canada (Attorney General), [1930] AC 124, a.k.a. the Persons Case, and quoted by Lord Hewart, Lord Sankey emphasized the importance of the Rule of Law, and of the courts as its enforcers:

Amid the cross-currents and shifting sands of public life the Law is like a great rock upon which a man may set his feet and be safe, while the inevitable inequalities of private life are not so dangerous in a country where every citizen knows that in the Law Courts, at any rate, he can get justice. (151)

And then, describing the threats to the courts’ role in upholding the Rule of Law, Lord Sankey pointed to

what has been described as a growing tendency to transfer decisions on points of law or fact from the Law Courts to the Minister of some Government department. (151)

And as for Lord Hewart himself, he did have an answer to at least one objection to judicial oversight of the administrative state that the defenders of deference still trot out from time to time: that allowing unobstructed judicial review of administrative decisions will lead to too much costly litigation. (For instance, in Edmonton (City) v Edmonton East (Capilano) Shopping Centres Ltd, 2016 SCC 47, [2016] 2 SCR 293, Justice Karakatsanis’ majority opinion claimed that “[a] presumption of deference on judicial review … provides parties with a speedier and less expensive form of decision making”. [22]) Lord Hewart responded to this concern by pointing out that

what is desired is not that there should be endless litigation but rather that litigation should be rendered as a rule unnecessary by the diffused and conscious knowledge that, in case of need, recourse might be had to an impartial public tribunal, governed by precedent, and itself liable to review. (155)

The point is one that goes to the very nature of the Rule of Law:

Nobody outside Bedlam supposes that the reason why Courts of law exist in a civilized community is that the founders of the State have believed happiness to consist in the greatest possible amount of litigation among the greatest possible number of citizens. The real triumph of Courts of law is when the universal knowledge of their existence, and universal faith in their justice, reduce to a minimum the number of those who are willing so to behave as to expose themselves to their jurisdiction. (155)

Just last year, the UK Supreme Court adopted essentially this reasoning in R (Unison) v Lord Chancellor [2017] UKSC 51, in the course of explaining the importance of access to adjudication ― perhaps ironically, in that case, adjudication in administrative tribunals, albeit ones functioning quite differently from those decried by Lord Hewart. Arch-anti-administrativist he may have been, but Lord Hewart was a more intelligent, and is a more relevant, jurist than those who dismiss him might realize. If you are interested in administrative law and haven’t read The New Despotism, you probably should read it.

Precedent and Democracy

“Long-standing” precedent is generally regarded as more authoritative than one of recent vintage. But there is reason to question that assumption, too. The more ancient a rule, the more likely it is that the reasons that made it sensible or good (whatever one’s criteria for the goodness of legal rules!) at the time it crystallized or was laid down no longer hold true. In the extreme case, we are left with the situation that Oliver Wendell Holmes famously decried in “The Path of the Law,” 10 Harv. L. Rev. 457 (1897):

[i]t is revolting to have no better reason for a rule of law than that so it was laid down in the time of Henry IV. It is still more revolting if the grounds upon which it was laid down have vanished long since, and the rule simply persists from blind imitation of the past.

So which sort of precedent deserves greater deference ― old or new? One consideration that tends to be missing from the debates about the authoritativeness of precedent ― even though it is a popular argument in the broader debate about the authoritativeness and legitimacy of “judge-made” (or, to borrow a less loaded term from Lon Fuller, “adjudicative”) law ― is legislative acquiescence.

It is often said that adjudicative law is democratically legitimate because, even though the courts in whose decisions it is set out are not in a meaningful sense democratic, legislatures could change the rules of adjudicative law that they do not like. Their failure to act is regarded as a sign of consent to the rules set out by courts, a democratic confirmation, albeit a tacit one, of the suitability of these rules for the community. Now, the acquiescence thesis does not strike me as entirely plausible, because the mass of case law is such that legislatures seem to me not very likely to be aware of all of its developments, still less to have the opportunity to respond to them in a timely fashion. But there is at least some truth to it, even if not enough to make it the conclusive argument for the legitimacy of adjudicative law its many proponents think it.

If we accept the acquiescence thesis, it would seem that long-standing precedents do have more authority than recent ones. The older a precedent, the more plausible the claim that the legislature has acquiesced to it.  The longer a precedent has been around, and especially the more subsequent judicial decisions have relied on it (admittedly, not necessarily a perfectly correlated fact), the more likely it seems that the legislature will become aware of it. People who stand to be affected by it and who are unhappy about the situation will, presumably, at least try to interest the legislature in their plight. And, given enough time, the legislature might respond.

But now, consider a somewhat stronger version of the acquiescence thesis. This stronger version holds that legislatures do not merely acquiesce to the rules of adjudicative law, but actually, albeit again implicitly, delegate rule-making responsibilities to courts, in more or less the same way  as they delegate such responsibilities to the executive branch of government. Again, I do not find this an entirely persuasive claim; I’m not sure that legislative silence on a certain point can really be taken as an invitation for the courts to deal with it rather than a merely lack of attention or even a deliberate decision not to regulate. Still, again, the delegation thesis is at least sometimes true. Legislatures do enact very general, even vague, statutes which they expect the courts to elaborate into more detailed regulatory schemes. And perhaps legislatures have in fact a more general expectation that if a problem arises with the law, the courts will deal with it ― it is hard to tell.

But if, or to the extent, that the delegation thesis is true, the courts should be quite proactive in responding to changing social conditions. They should then also be more suspicious of, rather than more deferential to, older precedents. The reasons that justified the precedent may have disappeared with the changes in society, the growth of our knowledge (scientific, economic, etc.), or even the development of other areas of the law.

Note, by the way, that the acquiescence and the delegation thesis are actually very close. The latter is only a mildly stronger version of the former. Indeed the delegation thesis depends on the acquiescence thesis for whatever normative validity it might have. If the legislature is not actually in a position to review and either consent to or revise adjudicative law, then it seems quite wrong for it to delegate legislative power to courts. (Though it might still be wrong for it do it even if the acquiescence thesis holds ― a point for a separate post, perhaps.) The difference between acquiescence and delegation seems to be only a matter of perspective: does the legislature consider the work of courts beforehand or retrospectively? In fact, to the extent the legislature does consider adjudicative law at all, it seems plausible that it does both, looking at what courts have done on an issue in the past and at what they might do in the future.

And yet, when it comes to the effect of time on authority of a precedent, the acquiescence and the delegation thesis seem to suggest opposite conclusions. I am not sure what to make of all this.

Less Than Meets the Eye?

Last week, the BC Court of Appeal issued what seems to me an important decision upholding the constitutionality of a treaty between the Nisga’a Nation, British Columbia, and Canada, and legislation implementing the treaty. The appellants in Sga’nism Sim’augit (Chief Mountain) v. Canada (Attorney General), 2013 BCCA 49, argued that the treaty and the legislation effected an irrevocable delegation of legislative powers to the Nisga’a government, effectively creating a third order of government, contrary to the constitution, which recognizes only two, Canada and the provinces.

As the Court describes it,

the Treaty has four basic components: first, a grant of 2,000 km² of land in fee simple to the Nisga’a Nation in substitution for Aboriginal title; second, the definition of hunting, fishing and trapping rights in Nisga’a lands extending beyond the land granted in fee simple; third, financial compensation to the Nisga’a Nation in consideration of the settlement of their historic claims; and fourth, the establishment of a Nisga’a government, with legislative powers in respect of defined subject matters, in accordance with a Nisga’a Constitution. (Par. 13)

And the treaty provides that the legislation enacted by the Nisga’a in some specified areas will prevail over inconsistent federal and provincial legislation. In other areas, it’s the other way around.

By its own terms, the treaty confirms and confers rights that come within the scope of s. 35 of the Constitution Act, 1982, and which thus enjoy constitutional protection. The appellants argued that this made the treaty and the legislation implementing it irrevocable, and the delegation of powers from Parliament and the BC legislature to the Nisga’a government permanent. That, they said, was a violation of the constitutional principles limiting delegation, pursuant to which a delegation of legislative power must be revocable, and the delegate must remain subordinate to the delegating legislative body. It also set up an legislative body not contemplated by the constitution, to which Parliament and the legislature have abdicated some of their legislative powers, in contravention of another constitutional principle.

The Court rejected these claims, holding that the treaty did not have the far-reaching effect ascribed to it by the appellants. Its grants of power could always be revoked, despite s. 35 of the Constitution Act, 1982, because the rights that it protects can be limited or infringed, so long as the infringement is justified (in a manner very similar to the justification of infringements of Charter rights under s. 1 of the Charter):

 In the result, government retains the right to pass legislation or take action that infringes the s. 35 rights recognized by the Treaty.  If government were to do so, the Nisga’a Government would be entitled to challenge the validity of the legislation or government action.  In the face of that challenge, government would be required to justify the infringement of the Treaty right …  If government were successful, the infringement would be upheld.  … [N]othing in the Treaty prevents matters from unfolding in this way. (Par. 78)

For the same reason, says the Court, the abdication argument must fail. Since Parliament and the legislature can infringe the treaty, provided the infringement is justified, they can in theory legislate over the matters on which the treaty purports to make Nisga’a law supreme. In addition,

the model of government established by the Treaty recognizes concurrent jurisdiction supplemented with prevailing-law rules.  This model necessarily contemplates the possibility of conflicts or inconsistencies between Nisga’a laws and federal or provincial law.  The Treaty defines what occurs in the event of conflict or inconsistency, balancing Nisga’a, federal and provincial interests.  The Treaty does not confer exclusive law-making powers on the Nisga’a Government.   (Par. 86)

(The appellants also made a number of other arguments, but they were long shots, and the Court made rather short work of them.)

This decision seems somewhat similar to the fixed election date case, Conacher v. Canada (Prime Minister), 2010 FCA 131, in which the Federal Court and the Federal Court of Appeal held that the impugned provisions were, in effect, meaningless, in order to avoid holding that they were unconstitutional. Yet there is also a difference, in that the Court acknowledges that the treaty has at least some effect: it entrenches the delegation of legislative power to the Nisga’a in all cases where an infringement on that delegation cannot be justified.

So I don’t think that, even on the Court’s approach, it is fair to say that the delegation of legislative power in no way removes the Nisga’a government from the position of subordination to Parliament and the provincial legislature. And, of course, this approach seems rather disingenuous. Surely, the point of the treaty is to make sure that there will be no infringements of the rights that it confirms. The Court says there is less to the treaty than meets the eye, but makes itself guilty of Nelsonian blindness in order to come to that conclusion.

It seems to me that the better way to proceed would have been to engage with the difficult issue of whether the limits on delegation and abdication apply to delegations of legislative powers to First Nations. There were some arguments to the effect that First Nations have an inherent right to self-government that was enough to justify the treaty, but the Court chose not to answer them, thinking that this was not necessary and, therefore, not prudent. It tried to choose the safe solution, but in my view, it is at best an inelegant, and at worst a seriously flawed one.