Just Hook It to My Veins

Judge Amy Coney Barrett’s excellent lecture on statutory and constitutional interpretation

Justice Scalia’s 1989 Lecture on “Assorted Canards of Contemporary Legal Analysis” is well known; indeed it has featured in a post by co-blogger Mark Mancini. Judge Amy Coney Barrett of the US Court of Appeals for the 7th Circuit revisited that lecture last year, and her remarks have been published recently, as “Assorted Canards of Contemporary Legal Analysis: Redux“. They are a short and profitable read, including for Canadian lawyers, to whom almost everything Judge Barrett says is relevant. Judge Barrett’s comments have mostly to do with statutory and constitutional interpretation, but they also touch on the issue of “judicial activism”. And I agree with just about every word.

The main topic Judge Barrett addresses is textualism. She defines it as the approach to interpretation that

insists that judges must construe statutory language consistent with its “ordinary meaning.” The law is comprised of words—and textualists emphasize that words mean what they say, not what a judge thinks that they ought to say. For textualists, statutory language is a hard constraint. Fidelity to the law means fidelity to the text as it is written. (856; footnote omitted)

Judge Barrett contrasts textualism, so understood, with the view that “statutory language isn’t necessarily a hard constraint. … Sometimes, statutory language appears to be in tension with a statute’s overarching goal, and … a judge should go with the goal rather than the text”. (856) Judge Barrett labels this latter approach “purposivism”, but that is perhaps not ideal, since many approaches to interpretation and construction, including ones that respect the primacy and constraint of the text, might properly be described as purposive.

With this in mind, the first three of Judge Barrett’s canards are “textualism is literalism” (856); “[a] dictionary is a textualist’s most important tool” (858); and “[t]extualists always agree”. (859) She notes that

[l]anguage is a social construct made possible by shared linguistic conventions among those who speak the language. It cannot be understood out of context, and literalism strips language of its context. … There is a lot more to understanding language than mechanistically consulting dictionary definitions. (857)

The relevance of context to interpretation is an important reason why textualists (and originalists) don’t always agree. If it were simply a matter of consulting the dictionary and the grammar book,

one could expect every textualist judge to interpret text in exactly the same way. Popping words into a mental machine, after all, does not require judgment. Construing language in context, however, does require judgment. Skilled users of language won’t always agree on what language means in context. Textualist judges agree that the words of a statute constrain—but they may not always agree on what the words mean. (859)

The example of such disagreement that Judge Barrett provides concerns the interpretation of the provisions of the US anti-discrimination statute ostensibly directed at discrimination “because of sex” as applying, or not, to sexual orientation and gender identity. Judge Barrett describes what happened at the US Court of Appeals for the 7th Circuit, but a similar disagreement arose when the matter was decided by the Supreme Court in Bostock v Clayton County, 140 S.Ct. 1731 (2020). Mark wrote about it here.

Judge Barrett then turns to another issue, this one concerned specifically with constitutional interpretation: should the constitution be interpreted differently from other legal texts? The idea that it should ― for which Chief Justice Marshall’s well-known admonition in McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316 (1819) that “we must never forget, that it is a constitution we are expounding” (407) is often taken to stand ― is Judge Barrett’s fourth canard. For her, “the Constitution is, at its base, democratically enacted written law. Our approach to interpreting it should be the same as it is with all written law.” (862) To be sure, the Constitution contains “expansive phrasing and broad delegations of congressional and executive authority to address unforeseen circumstances”. (862; footnote omitted) (One might do an interesting comparison of the US and Canadian constitutions on this point: as a very superficial impression, I am tempted to say that the delegations of legislative power are more precise in the Constitution Act, 1867, but those of executive power are even more vague.) But while constitutional language differs from that of an ordinary statute, the ways in which it should be interpreted do not: “[t]he text itself remains a legal document, subject to the ordinary tools of interpretation”. (862) Indeed, as Justice Scalia already argued, this the only reason for having installing courts as authoritative interpreters of the Constitution: were it not an ordinary law, why would we allow ordinary lawyers to have anything to do with it?

In particular, the principle that “the meaning of the law is fixed when it is written”, which is “a largely, though not entirely, uncontroversial proposition when it comes to statutory interpretation”, (863) applies to the Constitution too. This principle is indeed recognized, in statutory interpretation, even by the Supreme Court of Canada: R v DLW, 2016 SCC 22, [2016] 1 SCR 402 is a recent example. In the constitutional realm, however, our Supreme Court buys into the canard denounced by Judge Barrett ― or at least says it does. (Reality is often different.) As Judge Barrett explains, “as with statutes, the law [of the Constitution] can mean no more or less than that communicated by the language in which it is written” (864) ― and what that language communicates must of course be understood with reference to what it meant when it was communicated, not what it would come to means at some future date.

Judge Barrett makes an additional point which requires some clarification in the Canadian context, so far as statutes are concerned. It concerns the importance of compromise to the drafting of legal texts. For Judge Barrett, since laws reflect arrangements reached by representatives of competing or even conflicting interests, their interpreters should seek to give effect to these agreements and compromises, notably through “reading the text of the statute at the level of specificity and generality at which it was written, even if the result is awkward”, (863) and even when it might seem in tension with the statute’s purpose. The Canadian caveat is that our statutes are, at least to some extent, less the product of compromise than those of the US Congress. Especially, but not only, when they are enacted by Parliaments and legislatures where the executive has a majority, they reflect the executive’s policy, and are primarily drafted by officials executing this policy. But one should not make too much of this. As I pointed out here, statutes ― including in Canada ― often reflect compromises between a variety of purposes and values, even if these compromises are the product of a cabinet’s disucssions or even of a single politician’s sense of what is right and/or feasible. It follows that statutes should indeed be read carefully, with text rather than any one among these purposes being the interpretive touchstone. And as for constitutional interpretation, Judge Barrett’s point applies with full force. It was nowhere better expressed than by Lord Sankey in the  Aeronautics Reference, [1932] AC 54, [1932] 1 DLR 58:

Inasmuch as the [Constitution Act, 1867] embodies a compromise under which the original Provinces agreed to federate, it is important to keep in mind that the preservation of the rights of minorities was a condition on which such minorities entered into the federation, and the foundation upon which the whole structure was subsequently erected. The process of interpretation as the years go on ought not to be allowed to dim or to whittle down the provisions of the original contract upon which the federation was founded, nor is it legitimate that any judicial construction of the provisions of ss. 91 and 92 should impose a new and different contract upon the federating bodies. (DLR, 65)

After briefly passing on the idea that “judicial activism is a meaningful term” (865) ― she notes that, actually, “there is no agreed-upon definition of what it means to be an activist” (865) ― Judge Barrett turns to the view that a legislature’s failure to overrule a judicial decision interpreting an enactment can be taken as assent to the interpretation. For one thing, since the meaning of a statute is fixed at the time of its enactment, “what a later Congress” ― or Parliament or legislature ― “thinks is irrelevant”. (867; footnote omitted) But further, “even if we did care, there is no way to reliably count on congressional silence as a source of information”. (868) Silence might just mean that the legislature is unaware of the decision, or it might mean that the legislature finds intervention inexpedient, or not enough of a priority, though desirable in the abstract. Judge Barrett does not quote Sir Humphrey Appleby, but she reminds us that we ought not to mistake lethargy for strategy. Judge Barrett also refers to the bicameralism-and-presentment legislative procedures of the US Constitution, but that discussion is probably less relevant to Canadian readers.

Indeed I am not sure how salient this issue of acquiescence-by-silence is in Canada, as a practical matter. I don’t seem to recall decisions invoking this argument, but I may well be missing some. Judge Barrett’s attention to it is still interesting to me, however, because it is one of the possible justifications for the persistence of adjudicative (or as Bentham would have us say “judge-made”) law (not only in statutory interpretation but also in common law fields) in democratic polities.

In that context, I think that Judge Barrett is right that we cannot draw any concrete inferences from legislative silence. My favourite example of this is the issue of the admissibility of evidence obtained in “Mr. Big” operations, where suspects are made to believe that confessing to a crime is the way to join a powerful and profitable criminal entreprise. Such evidence was largely admissible until the Supreme Court’s decision in  R v Hart, 2014 SCC 52, [2014] 2 SCR 544, which made it presumptively inadmissible, except when tight safeguards are complied with. This was a major change, framed in almost explicitly legislative language. Yet Parliament ― with, at the time, a majority ostensibly focused on law-and-order issues ― did not intervene in response to the Supreme Court’s decision, just as it had not intervened before it. Does this mean Parliament agreed with the law as it stood before Hart, and changed its mind as a result of Justice Moldaver’s reasons? Probably not. What does its silence mean, then? Who knows. As Judge Barrett suggests, this does not really matter.

I wouldn’t have much to write about, I suppose, if I always agreed with the courts. I should be more grateful than I tend to be to judge who make wrongheaded decisions ― they may be messing up the law and people’s lives, but they are helping my career. Still, at the risk of depriving myself of future material, I call upon Judge Barrett’s Canadian colleagues to read her remarks and to take them on board. They are smart and show a real commitment to the Rule of Law. And, on a more selfish note, it really is nice to agree with a judge for a change.

Animated GIF - Find & Share on GIPHY

Contrarians at the Gates

On responsible scholarship and engagement with heterodox ideas

Professor Newman has posted his own response to the “article” in which Stepan Wood, Meinhard Doelle, and Dayna Scott attempt to besmirch his well-earned reputation as one of Canada’s leading constitutional law scholars. As he says, “it is a serious threat to responsible academic discourse to make mistake-riddled arguments in ways that paint academic interlocutors as personally irresponsible and lacking integrity”, which, as he shows, is what Professors Wood, Doelle, and Scott have done. As he also explains, while his critics purport to be concerned about “responsible scholarship”, their argument is so focused on just one article that he wrote that one cannot but “ask what the goal really was”. I will venture some speculation about the answer to this question.

To be blunt, I think that the goal is to constrain the scope of what may be said by academics writing on politically salient issues, how it can be said, and where. Now, Professors Wood, Doelle, and Scott deny this. They write:

Vigorous debate and disagreement are the lifeblood of academic discourse and the engine for advancement of knowledge. To insist on rigour and fairness in such debate is not to impose “political correctness” on scholars who espouse unpopular views. Nor is it a manifestation of the fragility of a liberal academic establishment unable to handle controversial perspectives. (13)

Of course this is true, as a general proposition. But we need not take on faith the claims of those who would have us believe that they do no more than dispassionately insist on rigour and fairness. We can look at the specifics of their argument, and at the way in which it is framed. Professor Newman has mostly done the former. I shall mostly do the latter.

One striking thing about the Wood, Doelle, and Scott article is that it is not just about “responsible scholarship” as a timeless value, as Professor Newman’s response is. It’s about “responsible scholarship in a crisis” (emphasis mine). This framing is not just a flourish. The entire opening section of the article argues that debates about climate policy are occurring “in the context of an unprecedented crisis”, (2) and that scholarly commentary ― and hence the norms of responsible scholarship ― are especially salient at such times because “[a]ctors in government, civil society and business often appeal to academic expertise to diagnose and resolve crises”. (3)

The norms to which Professors Wood, Doelle, and Scott appeal are not crisis-specific, to be sure (though, as Professor Newman shows, their relevance to their argument is questionable), but this framing is not innocent. It reinforces the dynamic which my colleague Allan Beever decries in his article on “Engagement, Criticism, and the Academic Lawyer”, (2017) 27 New Zealand Universities Law Review 1111. Professor Beever suggests that

academic lawyers, given their subordinate position in the legal system, are all too desperate to believe that they really matter, believing that if they matter that must be in something like the way that judges matter, thereby coming to believe that “dangerous” ideas have to be battled against in the way that they would be were they, say, influencing the decisions of the Supreme Court. (1125)

Of course, if the dangerous ideas are likely to influence the decision of a Supreme Court in a crisis, they are all the more dangerous and must be battled against all the stronger. This is why I say that Professors Wood, Doelle, and Scott are seeking to limit what can be said about politically salient issues: it is the topical nature and valence of Professor Newman’s ideas that trigger their attack. And they make no bones about the fact that they indeed worried about Professor Newman’s ideas influencing the Supreme Court of Canada’s consideration of the constitutionality of the federal carbon tax legislation.

I suspect, moreover, that this “crisis” framing helps explain why Professors Wood, Doelle, and Scott escalate the battle further and attack not only Professor Newman’s “dangerous” ideas but his integrity. His work, they say, “crosses a line that separates distortion and disparagement from constructive scholarly debate” (12) and “does not uphold standards of scrupulous fairness in scholarly research”. (13) It cannot be trusted and should be summarily disregarded. This makes actual scholarly debate unnecessary and indeed impossible ― there’s no point or even meaning in debating a dishonest person.

Yet this is facile and self-serving. Instead of doing the hard work of refuting the arguments they disagree with, Professors Wood, Doelle, and Scott attempt to discredit the person advancing them. This is also, of course, a myopic tactic that can and is bound to be used against scholars who agree with the substantive positions that Professors Wood Doelle, and Scott would defend. Indeed I’m pretty sure that ad hominem attacks on academics have long been more of a “thing” on the political right than on the left, not necessarily because the right is somehow even more immoral than the left, but simply because the academy has always tilted leftwards, making it easier for the right to find targets there. The tilt is growing ever stronger, and as it does so the cost to the political right of attacking not only individual scholars but also entire disciplines and the academy as a whole falls ― there is less and less of a risk of making victims with friendly fire. “Progressive” scholars who make personal discreditation an acceptable way of conducting academic disputes are only helping sharpen the weapons that will be directed at them.

But regardless of political implications, what is certain is that personal discreditation destroys the possibility of genuine scholarly debate bringing truth to light. Such debate requires mutual criticism, but is incompatible with enmity. As Professor Beever writes, “it is important … to distinguish between criticism and hostility. Criticism can be extremely robust, but it always takes its object seriously.” (1125) Professors Wood, Doelle, and Scott don’t want their readers to take the work they attack seriously, and so prove Professor Beever’s point that “[i]n law … hostility all too frequently prevents genuine criticism from occurring”. (1125)

Let me turn now briefly to my claim that Professors Wood, Doelle, and Scott want to limit the manner in which scholarly debate can happen. They chide Professor Newman both for criticizing the work of particular scholars, for example making a point of noting that one of them is “his junior untenured colleague”. (Ironically, especially given Professors Wood, Doelle, and Scott’s concern with distortions, the person in question is no longer Professor Newman’s colleague.) But then they also accuse Professor Newman of “casual generalizations [that] are examples of sloppy research”, (10) because he does not name other scholars who represent the trends against which he inveighs.

You’re damned if you do name specific scholars to criticize ― especially if they are junior colleagues ― and equally damned if you don’t. Now, one might debate how much a scholar should name the names of those he or she criticizes. Professor Beever suggests that this should be done sparingly, and widely held positions should be attacked without singling out individual representatives. I’m not sure I quite agree. But in any case it’s one or the other. Either generalizing is bad, and one should focus on individuals ― but then, it really shouldn’t matter who they are ― or one should be allowed to generalize. (I’ll note that, as a still relatively junior academic, I rather resent the suggestion that I should be treated with kid gloves by my elders and betters, at my institution or anywhere else.)

Besides, there is also something perverse about Professors Wood, Doelle, and Scott attacking the tone of Professor Newman’s article and its alleged lack of fairness to the targets of its criticism in an article that is anything but charitable, and indeed quite unfair, to Professor Newman, as he eloquently shows. Yet this is hardly surprising. There is a consistent asymmetry to tone-policing in the Canadian legal community. Scholars who criticize the received wisdom and its upholders are expected to be on their best behaviour. Those who uphold it are held to no such standard.

Finally, I turn to the question of where scholarly debate can take place. Professors Wood, Doelle, and Scott write that “[a] rigorous peer review process would normally catch most problems like the ones we have identified”, (13) and add, in a footnote, that “[t]he Saskatchewan Law Review failed to reply to inquiries … whether Professor Newman’s article was peer reviewed”. (13) Some of the subsequent Twitter discussion also focused on this issue. Yet to the extent that the implication here is that “responsible scholarship” is peer-reviewed scholarship, this is beside the point.

One issue is that, just like the invocation of crisis and the call for respectful tone, the appeal to the authority of peer review is less innocent than it might seem. As I wrote here, “the peer review process is a bit of a crapshoot” for heterodox ideas. Even when they are well argued, “some reviewers will bristle and see their role as that of gatekeepers preserving scholarship from heresy”. To be sure, as I further pointed out, heterodox scholarship can make it through peer review. But the issue of excessive gate-keeping, often applying double standards, is real enough. To insist that only scholarship that has made it through peer review matters is to load the scales in favour of conformist ideas, which have an easier time overcoming this hurdle.

But the real problem is more fundamental. Peer review simply does not guarantee quality; plenty of rubbish makes it through peer review and gets published, while good ideas get rejected. When Joshua Gans and George Shepherd asked

140 leading economists, including all living winners of the Nobel Prize and John Bates Clark Medal, to describe instances in which journals rejected their papers [they] hit a nerve. More than 60 percent responded, many with several blistering pages. (165)

It would be interesting to see a similar study in law, but I rather doubt that peer review in our discipline does much better.

Conversely, the fact that an idea did not go through peer review ― for example because it was published in an American journal, or even on a blog ― does not mean that it is bad. Readers can decide for themselves. A lawyer, or at least an expert in a given field, let alone a judge assisted by a platoon of clerks, can always check for him- or herself whether an argument holds up ― whether the sources it cites support it, whether it is missing something. If one wants to criticize the defects of an argument, one should identify them, instead of lazily musing about whether the argument has been peer reviewed.

This, ultimately, is a big part of what the matter comes down to: faced with deep and seemingly consequential disagreements, are we willing to do the hard work of explaining why our opponents are misguided and mistaken? Or are we content to discredit or tone-police them, or say that they didn’t published their ideas in the right format or in the right venue, so as to avoid substantive engagement?

But the issue is not limited to the avoidance of hard work. In a crisis it is tempting to take shortcuts. The lure of sophistry is too powerful to resist. If ad hominem attacks, tone policing, and arguments from authority can help defeat the danger that wrong ideas, or wrong people, will influence decision-makers, why not resort to them? The contrarians are at the gates ― this is no time for old-fashioned notions of probity.

Yet probity, as well as curiosity about ideas one disagrees with, and friendliness to those who expound them, are the perennial values that no crisis can put to rest. Responsible scholars, tear down these walls! Open these gates!

On John Willis and the Pesky Politics of Administrative Law

John Willis was and is considered one of Canada’s most important administrative law academics. As a student of administrative law and the law of judicial review, one cannot skip Willis’ classic works, like his books “The Parliamentary Powers of English Government Departments” and “Canadian Boards at Work”—and his caustic papers, including his attack on the McRuer Report and his famous “Three Approaches to Administrative Law: The Judicial, The Conceptual, and the Functional.” Especially in this latter piece, Willis sets out his comprehensive functional theory of the law of judicial review—that theory holds that courts, with a typically “conservative” orientation, could never understand the difficulties of governments that, post-New Deal, were concerned with social welfare. Accordingly, courts should butt out, in service of the expertise, efficiency, and progressive orientation of administrative decision-makers.

While Willis should obviously be commended and respected as an eminent scholar of administrative law, there is a core problem at the centre of Willis’ thought that should stand as a warning for us in the modern era. Willis fundamentally viewed administrative law as a project of politics. As R. Blake Brown notes in this article, and as Willis himself always argued, the law of judicial review (and administrative law more broadly) was not about legal principles or controls on the administrative state, but was rather designed to limit the interference that legalist, conservative courts could wrought on tribunals pursuing the social good in an expert way. But this sort of thinking runs into two fundamental problems: (1) it ignores the fact that, strategically, administrative power can and has been used to fulfill the policy aims of governments who do not have any designs on social welfare goals—this was a clear blind-spot in Willis’ own work, one that led him to over-trust government; and (2) normatively, as recent Supreme Court and Federal Court of Appeal jurisprudence tells us, there is a meaningful difference between law and politics when it comes to the interpretation and application of laws governing judicial review. Granting deference based on the supposed underlying political motivations of particular decision-makers undermines this separation.

Let’s start by reviewing Willis’ functionalism. Willis self-described himself as a “government man,” and “what actually happens man.” By this he meant he was less concerned about the legal principles of a 19th century constitution, but rather was concerned with the making of “effective government” (see “Administrative Law in Retrospect” at 227). Functionalism crafted doctrine to align with the way government operates and the programs government is responsible for implementing. At the time of Willis’ writing, the struggle for government was the implementation of social welfare programs, closely associated with the New Deal. Delegation to administrative tribunals was one of the ways that these programs were implemented. A functional approach, then, would respect the legislative choice to delegate

In my view, deference to administrative tribunals in the functionalist mould was supported by a number of presuppositions about administrative decision-making, but the most important one for our purposes was what I call the “political” presupposition. Deference on the functionalist account was justified because of the apparent political valence of the decision-makers under review and what they were designed to achieve. Says Michael Taggart (at 257), describing the functionalists of the era:

These left-leaning scholars were deeply resentful of what they saw as conservative judges twisting the pliable rules of statutory interpretation to favour the existing order, privileging the rich and the powerful, and defeating the purposes of statutes intended to further the interests of the workers, the homeless, and the least well-off in society.

Deference on functionalist grounds was therefore a reactionary force, one that was a political project designed to fight back against the supposedly conservative orientation of the courts, that used legal principles to stunt the social welfare programs of governments.

Willis himself clearly fell in this category. A social democrat, Willis railed against any sort of thinking that would interfere with the prerogatives of government, undertaking social programs. He viewed government as fundamentally changing in light of the New Deal:

The State had changed its character, had ceased to be soldier and policeman, and was rapidly becoming protector and nurse…Again the right of the community bulk larger than the rights of the individual (See Parliamentary Powers at 13 and 51).

Risk described Willis’ functionalism as such:

Willis’ thinking about law and government can best be summarized by dividing it into three parts. The first is his observations and attitudes about government and its institutions….He perceive the nature and extent of the expansion of government, and its implications for the structure and functions of the legal institutions. He perceived a changing relation between the individual and the community, and how legislative policies were expanding the claims of the community against the individual, and circumscribing common law ideals….He had a great faith in experts, and he believed the courts should give liberal scope to the agencies on review (see Risk, at 545).

The political appreciation of administrative tribunals as representing the needs and wishes of the “community” was a constant thread through Willis’ scholarship advocating for a judicial “hands-off” approach to decision-makers like labour boards.

So, what are the problems with such an approach? To my mind, there are two. First, Willis’ political approach to deference fails on its own terms: it fails, on any complete account, to actually reach an ideal of social justice. This is because deference itself has no political valence, and can be easily used to vindicate decisions of administrators that run counter to social justice. And secondly, on principle, a political approach to deference runs counter to our positive law and to the good reasons for it.

Let’s first tackle the issue of social justice and its connection to doctrines of deference. As I’ve written time and time again, using deference as a means to reward the  decision-makers we like –because we ascribe to them some political ideal—is an unprincipled and politically naïve way to view the law. Let’s start with the latter contention. Willis’ supposition seems to be that courts themselves cannot be trusted to uphold the purposes of ameliorative legislation because of their conservative orientation. But it is not axiomatically true—and in fact, it seems bizarre—to suggest that deference will always serve to advance social welfare principles. It does no such thing.

The development of the administrative state is simple proof of this. When Willis was writing, he made the near-sighted appreciation that deference supported the administrative state as it existed at that time. At that time, observers were mostly concerned with labour tribunals, who were seen as consummate experts in their craft. But Willis either did not predict that deference would and could also have to attach to tribunals he did not appreciate under his social welfare rubric. That is, and I have said before, there are other aspects of the administrative state that do not map so neatly onto any past or modern description of social justice. Prisons, an area of interest for me, come to mind—perhaps the place where administrative discretion, at least prior to the CCRA, was most unbounded. Another example, that of immigration decision-making,  also comes to mind. What, beyond brash politics, justifies treating these tribunals any differently?  If one believes in deference, how can one say that prisons are any less deserving of deference than any other decision-maker?

This does not strike me as a consistent approach based on social justice. Later in Willis’ career, this thought must have occurred to him. That is because, in his “Administrative Law in Retrospect,” Willis addressed the question of a number “fashionable cults” which, to him, were negatively affecting the prerogatives of government:

This is very interesting coming from someone who is committed to social justice. How could it be consistent, if one accepts some political justification for deference based on social justice, for Willis to deny prisoners a right to be heard? Why are prisoners less good, in Willis’ eyes, than unions before a labour board? Willis’ myopic conception of social justice was profoundly underinclusive, even on its own terms.

Another explanation of this oddity is that Willis was not committed to social justice at all. Rather, it is very likely and possible that Willis was indeed a “government man” in the most literal sense of the term. Any action that could offend a government prerogative, in his eyes, was abhorrent. So the “cult of the individual,” and prisoners, all serve to run against government, even if government offends social justice. But this stands inconsistently with the idea of social justice. Social justice, on any cohesive account, is not about empowering government for the sake of government. The problem is that government can act in ways that contradict the theory of social justice

The above point challenges Willis on his own social justice terms, but there is an external, doctrinal reason to be wary of Willis’ approach to deference. It is indeed true that the fight for deference in Canada is overlayed by considerations of politics. After all, the laws delegating power to decision-makers—or laws that work to limit the scope of power for these decision-makers—are passed against the backdrop of a legislature that is a partisan organ. But that is a separate matter altogether from the actual legal justifications for deference, which like the interpretation of statutes, should be a separate concern from politics. Luckily, our law recognizes this fact clearly, and does so for good reasons. Vavilov, for example, does not base deference on any good-faith presumptions about the expertise of decision-makers. Now, the very fact that the legislature delegated power—any power—to any decision-maker is a fact worthy of deference. The tool of delegation as the grounds of deference has the benefit of being agnostic as to how one can judge, politically, particular tribunals. And Vavilov itself (at paras 120-121) cautions against reverse-engineering doctrine to suit a desired outcome. Clearly, Willis’ political approach to doctrine (and the arguably political approach of the common law courts) run afoul of our current law, which erects a clear separation between deference as a doctrine and the political results of a deferential approach. Justice Stratas in the Federal Court of Appeal has made a similar point in the context of statutory interpretation and judicial review: see Williams, at para 48; Cheema, at para 74; and Hillier, at para 33.

While that is the state of the positive law, it is the positive law for a good normative reason. It is orthodox today in the academy that law cannot be meaningfully separated from politics. It is even true that some say that any attempt to do so is necessarily “reactionary” or “conservative.” But this contention does not take account of the different parts of law-making and interpretation, and the very purpose of law itself. As I mentioned above, it is of course true that laws reflect the political consensus of the legislature at the time they are passed. It would be wrong and overbroad to suggest that the making of law is or should be divorced from the political process: indeed, it is the function of our legislatures to make laws that, at least in theory, are undergirded by the support of a majority. However, this is a completely separate act from the interpretation of laws. Laws, in order to be consistent with at least one aspect of the Rule of Law, must be general rather than specific; and when a judge interprets a law, she does so to give the meaning to the text, context, and purpose of the law that is enacted on the page (even purpose, as I discuss here, is usually and ultimately guided by text). The task of interpretation of laws should not be governed by consideration of politics; of what this or that judge thinks of this or that tribunal. Should it be the case that judges grant deference because of their political views, we will go along way towards undermining our separation of powers between courts and legislatures, imperfect as it is in Canada.

Of course, it is impossible to say that politics will never enter the interpretive activity. But that is a different question altogether than how doctrine is constructed. Ideally, the way we theorize deference and interpretation should not be based on political musings; rather, theory should be based on the foundational principles of our legal order, including the choice of a legislature to delegate power and the core interpretive function of the courts. It might be orthodox to suggest otherwise these days, but in my view, the very purpose of law in the law of judicial review is to enforce the limits that legislatures themselves provide—no matter how wide or loose they are—on administrative decision-makers. It should not be the role of the courts, as Justice Stratas so eloquently says in the above-cited decisions, for courts to pick winners and losers based on politics.

Back to Willis. John Willis’ contributions to Canadian administrative law will live for the ages. But his approach to the law of judicial review should not be celebrated wholesale. Willis’ cardinal mistake was falling victim to the game started by the “conservative” common law courts. If it is true that those courts struck administrative decisions because they did not appreciate the social welfare function of those agencies, that runs counter to our governing law and the good reasons for it. But today, Willis is still celebrated; the common law courts are not. I think it is fair and appropriate to draw attention to the blindspots in Willis’ theory: his myopia regarding what he thought was “social justice”; and the specious attempt to import deference based on some political justification.

On Responsible Scholarship

A Reply to Stepan Wood, Meinhard Doelle, and Dayna Scott

Dwight Newman, QC, BA, JD, BCL, MPhil, DPhil, Professor of Law & Canada Research Chair in Indigenous Rights in Constitutional and International Law, University of Saskatchewan

Accusations of irresponsible scholarship are a serious matter, and they have an even graver dimension when they give the appearance of being framed and timed so as to attempt to interfere with academic contributions to a major public debate.  In this post, I address a recent paper by Stepan Wood, Meinhard Doelle, and Dayna Scott attempting to challenge my well-known carbon tax article.  I must express serious concerns with their characterization of my article.  I must also express that the publication of their paper threatens academic discourse through the intimidating effects it could have on scholars.

I explain these points here and am grateful for the opportunity to be able to publish a response promptly.  The Dalhousie Law Journal declined to grant me the opportunity to publish a reply alongside the paper in their forthcoming issue. 

In 2019, I participated in what was pitched to me as a “duelling-articles debate” in the Saskatchewan Law Review after the Saskatchewan Court of Appeal’s decision in the carbon tax case.  For my contribution, I published a 6000-word article arguing that the carbon tax faced more constitutional obstacles than many had initially expected.  This was an argument about constitutional law—I am on record elsewhere as regretting that the federal government undertook a constitutionally problematic design for its carbon tax, as I am supportive of environmental policies that could include properly designed carbon taxes.  Constitutions are not mere debris to be run over on the road of public policy but must be respected as the rules of the road.

Last week, Professor Stepan Wood of the UBC Faculty of Law tweeted out his co-authored August 2020 Centre for Law & The Environment working paper (released in a paper series edited by Professor Wood) that is now also a forthcoming article in the Dalhousie Law Journal.  While aspects of that article frame it as an intellectual discussion on standards of responsible scholarship, (and it is somewhat longer than my own article), it is nonetheless focused almost entirely on my article. 

Both the conclusion of the paper and Professor Wood’s subsequent tweets confirm that he aimed to publish it before the Supreme Court of Canada hearing in which my article might be discussed, seemingly to try to discourage the Court from drawing upon my work by arguing that it was “irresponsible” research.  I do believe that the paper Wood tweeted out fulfilled what it set out to do and unfortunately does not contribute in any positive way to legal academic discourse.  The Wood, Doelle, and Scott (Wood/Doelle/Scott) paper paints what are reasonable scholarly points as irresponsible research, but it does so by pulling much of what I said out of context and inaccurately.  I will mention just some examples here:

  1. On p 6, Wood/Doelle/Scott suggest that I must not have read beyond the title of an article in interpreting the author as having hoped for changes in the constitution to accommodate climate change policy.  As it turns out, I had read not only the title but the article itself, as would be apparent in the fact that I engage with legal arguments contained in that article.  On the point at issue, that other scholar’s article has multiple passages referring to the need to adapt the interpretation of the constitution in light of new realities, which I am free to refer to as arguments for constitutional change.
  2. On pp 6-7, Wood/Doelle/Scott inaccurately say that I “complain” of “incoherence” in a scholar’s work where I made no such claim. 
  3. On p 7, Wood/Doelle/Scott suggest I should have read a particular article.  I actually discussed that very article and showed how it had problems in its understanding of the relationship between the legal doctrines of POGG and interjurisdictional immunity.
  4. On pp 7-9, Wood/Doelle/Scott object to my characterization of an elitist strand in some environmental law scholarship.  I cited  an article on regulatory capture (and actually did not challenge the part of the article on regulatory capture, as Wood/Doelle/Scott imply I did), and that article concludes with a three-page discussion of how academics can take a larger role in guiding the democratic process on environmental issues.  Some may think that a good idea or even a conventional one.  I simply identify its elitist dimension in a footnote that bears on the context for argument about the carbon tax.  I use accessible terms, but my footnotes make very clear that there are scholarly works that readers can consult further.
  5. On p 9, when I discussed a number of popular media pieces asserting Saskatchewan’s carbon tax litigation had no chance, Wood/Doelle/Scott say that I incorrectly interpreted the reference of the pronoun “we” in one of these pieces.  Wood/Doelle/Scott say that “we” referred to all Canadians.  I had followed the apparent reference of the pronoun “we” in the immediately prior sentence as those litigating on climate change.  Wood/Doelle/Scott could be right, but the op ed was ambiguous and was just as consistent with my reading.  To use this singular example of a different reading of “we” to claim that I made “distort[ing] statements” about media pieces is absurd.
  6. The Wood/Doelle/Scott paper also suggests that I am “unfair” to the courts because I provide a “blinkered” account of the case law (p 10).  A short article in a duelling-articles debate cannot discuss every authority at length, nor is it meant to do so.  Although Wood/Doelle/Scott criticize me for not engaging extensively with the case law in a footnote where I raise the possibility that the national concern branch of POGG is not well grounded in legal precedent, my footnote ends by saying “a full examination of those arguments would exceed the permitted limits for this article”, thus acknowledging that I was simply raising a possible argument that could be considered elsewhere.  I do not think any courts are troubled by this.

There are other examples.  Given that the Wood/Doelle/Scott paper sought to contribute to discussing the topic of responsible scholarship, it is difficult to see its contribution to that topic when it contains a significant number of errors about my article and what I say. 

A good article on responsible scholarship in a legal context could be quite interesting.  In their paper, at pp 4-5, Wood/Doelle/Scott link several quotes about norms of responsible scholarship, which might be the start of a discussion.  However, they take most of these quotes out of context, without explanation of the fact that they have done so.  In their proper contexts, these statements refer to approaches to data in scientific contexts or to internal operations of research teams.  Wood/Doelle/Scott apply them without discussion or explanation to treatment of scholarly sources and external scholars.  It would be interesting to analyze to what extent the pertinent norms for these contexts do or do not track precisely those other norms.  Discussing that would require an intellectual work going beyond what Wood/Doelle/Scott did.  In addition to considering the differences in the context of legal research, it would be important to consider the ways in which legal scholars write in a variety of different genres, ranging from more specialized academic contexts to more accessible practitioner-friendly pieces to public-facing pieces.  A strong contribution on responsible scholarship would also discuss and compare several different examples rather than focusing solely on one article combined with a few lines about another.  The Wood/Doelle/Scott paper falls short on these measures in a number of ways, leading one to ask what the goal really was.

It is one thing to make unconvincing arguments, as happens in both peer-reviewed and non-peer-reviewed work.  But it is a serious threat to responsible academic discourse to make mistake-riddled arguments in ways that paint academic interlocutors as personally irresponsible and lacking integrity.  Obviously, I agree entirely that responsible research practices facilitate good research and the contribution that academia can make to society.  However, purporting to adjudicate responsible academic discourse in the way that Wood/Doelle/Scott did sets a horrible precedent.  A scholar who makes an argument, popular or unpopular, should be able to do so without lightly being called irresponsible.  Other scholars who have seen events unfolding even in this one instance might well feel intimidated from participating in academic discourse by the prospect that they will face attacks not just on their ideas but on their integrity.


Ontario’s Superior Court strikes down the anti-carbon tax-sticker law, but still doesn’t get freedom of expression

Last year, I wrote about Ontario’s Federal Carbon Tax Transparency Act, 2019 (the “Act”) ― both about the disgraceful way in which it became law and about its unconstitutional speech compulsion, which I argued should not even be considered as a potentially justified limitation of the freedom of expression under section 1 of the Canadian Charter of Rights and Freedoms, because it was tantamount to the imposition of an official ideology. The constitutionality of the Act was in fact challenged by the Canadian Civil Liberties Association and, last week, in CCLA v Ontario (Attorney General), 2020 ONSC 4838, the Superior Court of Ontario struck it down.

At first glance, this is a welcome development for the freedom from compelled speech. Not only is the compulsion invalidated, but Justice Morgan’s approach might seem to bear some resemblance to the one I had proposed: in effect, he denies the government the chance to justify the Act under section 1. But look at Justice Morgan’s reasons more closely, and they turn out to be very narrow. Indeed, they could be used to support significant speech compulsions in the future.

This is not altogether surprising. Justice Morgan was constrained by the Court of Appeal’s reasoning in McAteer v. Canada (Attorney General), 2014 ONCA 578, 121 OR (3d) 1, upholding the constitutionality of the requirement that applicants for the Canadian citizenship swear an oath to bear “true allegiance” to the “Queen of Canada”, which I have described as a “parade of horribles“. And indeed it was none other than Justice Morgan who had written the first instance decision in that case. While it wasn’t quite as bad as that of the Court of Appeal, it did not evince much understanding of the harms of compelled speech either.

The Act required all gas stations to display a prescribed sticker alerting customers to the amount of the “federal carbon tax” levied on the gas they were purchasing. The evidence adduced by the CCLA showed that it was meant as a not-so-subtle intervention in the 2019 federal election campaign, in which the Ontario government supported the anti-carbon-tax position of the federal Conservatives and opposed the pro-carbon-tax Liberals. This partisan dynamic is a key factor in Justice Morgan’s reasoning.

Before getting to the substantive issues, Justice Morgan must address the Attorney General’s objection to the CCLA’s standing to challenge the Act. As it turns out, the CCLA has tried to enlist actual gas stations as plaintiffs or co-plaintiffs, but none would come forward. Justice Morgan explains that “retailers, with a view to market forces rather than to politics and constitutional law, have been loath to participate in this case” due to its political valence. [40] But the record to which Justice Morgan alludes suggests that this is not quite accurate: politics, in the shape of a fear of regulatory retaliation, seems to have been a motivating factor too. Be that as it may, Justice Morgan grants the CCLA public interest standing to pursue the case.

He must next decide whether the sticker requirement limits the freedom of expression protected by section 2(b) of the Charter. To this end, he applies the test set out in the Court of Appeal’s McAteer decision:

The first question is whether the activity in which the plaintiff is being forced to engage is expression. The second question is whether the purpose of the law is aimed at controlling expression. If it is, a finding of a violation of s. 2(b) is automatic. If the purpose of the law is not to control expression, then in order to establish an infringement of a person’s Charter right, the claimant must show that the law has an adverse effect on expression. In addition, the claimant must demonstrate that the meaning he or she wishes to convey relates to the purposes underlying the guarantee of free expression, such that the law warrants constitutional disapprobation. (McAteer, [69])

Justice Morgan finds that the sticker is indeed a form of expression. Yet in his view its purpose is not to control expression. In particular, he takes the view that “it would be difficult for the government to control expression by compelling certain messages … but not restricting others”. [50] Objectors remain free “to disavow” [52] the message they are compelled to voice, for example by posting disclaimers; hence their expression is not “controlled”. However, it is adversely impacted by the Act.

The key point for Justice Morgan is that, unlike the citizenship oath in McAteer, the sticker does not promote democracy and the Rule of Law. Indeed, it does not even serve to truthfully inform. Justice Morgan attaches some importance to the sticker’s use of the “carbon tax” nomenclature, which in his view is at odds with the Court of Appeal for Ontario’s opinion, in Reference re Greenhouse Gas Pollution Pricing Act, 2019 ONCA 544, 146 OR (3d) 65, that the policy at issue is not a “tax” within the meaning of the Constitution Act, 1867. Moreover, the sticker says nothing of the ways the money levied as carbon tax is distributed, in part to taxpayers, in part to provinces. As a result, it is a form of “spin”. [60] By requiring it, “the government is not so much explaining a policy [as] making a partisan argument”. [63] And “[b]y using law for partisan ends, the Ontario legislature has enacted a measure that runs counter to, rather in furtherance of, the purposes underlying freedom of expression”. [65]

This limitation of the freedom of expression is not justified under section 1 of the Charter. Indeed, unusually, Justice Morgan finds that the Act lacks a pressing and substantial purpose ― the first, and normally very low, hurdle a statute must meet to be upheld under section 1: “While truly informing the public about the components that make up the cost of gasoline would be a pressing and substantial purpose, promoting the Ontario governing party over the federal governing party is not.” [69] The Act is purely partisan rather than a real “policy choice”. [69] Justice Morgan goes through the other steps of the justified limitation analysis by way of an obiter, but it all comes down to his concern with partisanship. The Act is invalid.

Right outcome, but the reasoning is another matter entirely. Justice Morgan’s approach is illogical and conflicts with the Supreme Court’s precedents, notably inthat it collapses the two stages of the Charter analysis that the Supreme Court has always sought to keep distinct: first, the question of whether a right is being limited; second, that of whether the limitation is justified.

First, to say, as Justice Morgan does, that one’s expression is not controlled because one can disavow something one has been coerced to say is perverse. The fact that one is forced into disavowals shows sufficiently that what one is saying is not what one chooses to say.

The political context that Justice Morgan’s reasons depict highlights this problem. As he explains, it appears that gas station owners would rather keep quiet and sit out the political conflict about the carbon tax. This is their right ― the obverse of the freedom of speech is the freedom to stay silent. If they are forced into disavowals and denials, the gas stations will inevitably be taking sides in the political conflict they are trying to avoid ― if anything, this will be much more obvious than if they merely comply with the Act and display the required stickers. Of course, such a response is not what the Ontario legislature envisioned, but it would be caused entirely by the Act, and so it is absurd to deny that the Act amounts to a form of control of the gas stations’ expression.

Fortunately, the Supreme Court’s precedents mandate no such twisted inquiry. Ostensibly the most important freedom of expression case (I have argued here that it is only “leading from behind”), and the source of the “control” language used in McAteer and by Justice Morgan is Irwin Toy Ltd v Quebec (Attorney General), [1989] 1 SCR 927. The distinction between legislation that has control of expression as purpose and that which doesn’t is described as follows in the joint opinion of Chief Justice Dickson and Justices Lamer and Wilson:

If the government’s purpose is to restrict the content of expression by singling out particular meanings that are not to be conveyed, it necessarily limits the guarantee of free expression. If the government’s purpose is to restrict a form of expression in order to control access by others to the meaning being conveyed or to control the ability of the one conveying the meaning to do so, it also limits the guarantee. On the other hand, where the government aims to control only the physical consequences of certain human activity, regardless of the meaning being conveyed, its purpose is not to control expression. (974)

Applied to speech compulsions rather than censorship, this means that any legislation that “singles out particular meanings” that must be communicated, or forces an audience to listen to a communication, necessarily has control of expression as its purpose. Such legislation limits (or, as the Supreme Court often says, prima facie infringes) the freedom of expression. There is no need to consider effects, let alone to ask the purely subjective question of whether they are worthy of “constitutional disapprobation”.

This inquiry into effects and “disapprobation” in effect forces claimants to show that the law which compels their speech is not justified, and more specifically that it pursues an end worthy of judicial condemnation. The success of such an argument in this case should not blind us to the fact that this is a high hurdle. As noted above, this approach collapses the usual section 1 test of whether a limitation on a right is justified into the threshold inquiry of whether a right is limited in the first place, and it means that the claimant rather than the government bears the burden of proof. It follows that Justice Morgan’s streamlined approach to the section 1 analysis is rather less supportive of freedom of expression than one might think. The important work is already done by the time he gets there, as he has, in effect, found that the Act is unjustifiable. Had he not so found, he would have upheld it without ever getting to section 1, just as the Court of Appeal upheld the citizenship oath in McAteer.

Last but not least, Justice Morgan’s emphasis on partisanship as the fundamental problem with the Act is also misguided. For one thing, as tempting as it might be to say that partisanship can never be a sufficient justification for restricting Charter rights, the Supreme Court has in the past upheld laws that protect political incumbents from criticism, notably in Harper v Canada (Attorney General), 2004 SCC 33, [2004] 1 SCR 827. I think the Supreme Court was wrong to disregard the partisan valence of that legislation, but this shows that it will often be difficult to disentangle partisanship from other, specious considerations. Indeed, Justice Morgan himself suggests that a statute that is “a hybrid of policy and partisanship” would deserve to be treated as fully legitimate.

More importantly, Justice Morgan’s understanding of partisanship is very narrow. It does not encompass the promotion of a state ideology that rises, if perhaps only slightly, above the “horse race” version of partisan politics. He has nothing but sympathy for governmental “protection and promotion of Canada’s national and legal culture” [58] by means of forcing those who did not agree with this culture to voice loyalty to it. Needless to say, there is a political dimension to a “national and legal culture”, especially when this culture is coercively imposed by the state, even though Justice Morgan is oblivious to this. To him the distinctions between partisanship and high principle appear obvious. To the rest of us living in 2020, they are anything but.

Consider an obvious example: the late and unlamented “statement of principles” requirement that the Law Society of Ontario tried to impose on its members. Certainly its supporters argued in terms promoting a certain high-minded vision of social and legal culture (indeed they spoke of a “culture shift”). But then again, as we now know, there is a bitter partisan division over the issue within the ranks of the Law Society’s membership. So how would Justice Morgan approach the question of the constitutionality of the requirement? And would his approach be different now than it would have been before the partisan cleavage was revealed by the success of the StopSOP campaign in the 2019 Bencher election? Whatever we might think of the “statement of principles” or its opponents (of whom I was one), or of compelled speech more broadly, I hope we can agree that this is not a reasonable way of addressing such an important issue.

Of course it is a good thing that the Act is no more, and that the Ontario government, if it wants to continue its anti-carbon-tax propaganda campaign, will have to do it by itself, rather than by means of conscripting third parties. I have argued here that such ideological conscription is wrong when it serves to supposedly advance some rights-protecting agenda. It is no less wrong, obviously, when its aim has to do with fiscal and environmental policy. Governments have plenty of resources at their command. If they want to propagandize, they have no need to get unwilling individuals to do it for them.

Yet, the state of the law on compelled speech, and indeed on freedom of expression more generally, in Ontario at least, is cause for concern. It’s not just that few restrictions on freedom of expression are ever struck down. More importantly, the courts fail to understand what free speech means, and why it matters. Justice Morgan’s reasons for striking down the Act illustrate these failures just as much as his and the Court of Appeal’s earlier reasons for upholding the citizenship oath did.

Throwing Away the Key

Thoughts on life imprisonment without parole, in New Zealand and in Canada

Last week, Justice Mander of New Zealand’s High Court sentenced the Christchurch mosque shooter to life imprisonment without the possibility of parole for the murder of 51 people, attempted murder of 40 others, and terrorism. This punishment is provided for by section 103(2A) of New Zealand’s Sentencing Act 2002.

Justice Mander’s sentencing remarks in R v Tarrant, [2020] NZHC 2192hold some lessons for Canadians, as the Québec Court of Appeal is considering the appeals of both the Crown and the accused from the sentence the Superior Court imposed on the Québec mosque shooter in R v Bissonnette, 2019 QCCS 354. In that decision, about which Maxime St-Hilaire and I wrote here, Justice Huot found the possibility of stacking parole ineligibility periods for multiple murders in a way that amounted to sentencing those who commit them to life imprisonment without parole to a cruel and unusual punishment and a deprivation of liberty contrary to principles of fundamental justice, contrary to sections 7 and 12 of the Canadian Charter of Rights and Freedoms.

Although stating that “the needs of denunciation, of setting an example, and of incapacitation” are especially “pressing” [766; translation mine], Justice Huot went on to find that life imprisonment without a realistic prospect of parole was contrary to Canadian values. Canada, he wrote, “is not a land where the most undesirable elements of the community are shut in a gaol and their very existence forgotten, the key of their liberty having been thrown into the river of a vast collective indifference.” [845; tanslation Professor St-Hilaire’s and mine] For him, the possibility of rehabilitation, even for the worst offenders, means that it is “sophistry to assert that [multiple murderers] should reasonably expect, in a free, civilized, and democratic society, to spend the rest of their days behind bars”. [975] Justice Mander’s cogent remarks help show that this was wrong.

Justice Mander, it worth noting, is by no means insensitive to considerations of humanity and anti-populism that apparently influenced Justice Huot so much. He considers the prospects of rehabilitation, and notes that “[t]he sentence [he] impose[s] must represent a civilised reaction based not on emotion but justice and deliberation”. [177] But these concerns are not dispositive in a case such as this.

Addressing Mr. Tarrant, Justice Mander explains that his

prime objectives are threefold. First and foremost, to condemn your crimes and to denounce your actions. Second, to hold you accountable for the terrible harm you have caused — in plain terms, to attempt to impose some commensurate punishment … on behalf of the whole community, which in particular includes the victims of your crimes and their families, all of whom are a part of New Zealand’s multicultural society. Third … to protect the community from a person capable of committing cold-blooded murder on such a scale and who presents such a grave risk to public safety. [124]

Justice Mander notes that section 9 of New Zealand bill of Rights Act 1990 prohibits the imposition of “disproportionately severe … punishment” (judicially interpreted as calling for a test of gross disproportionality ― similar to the one applied to test the constitutionality of legislation under section 12 of the Canadian Charter of Rights and Freedoms). He notes, also, that “[t]here is European jurisprudence that indicates the imposition of a whole-of-life sentence in the absence of any effective review mechanism is incompatible with
international human rights instruments”. [139] Nevertheless, he finds that nothing short of a life sentence without parole would be proportionate to the crimes here.

Let me quote just one paragraph about the facts (this one drawn from Justice Mander’s discussion of the aggravating circumstances). It is horrible, and there is, alas, so much more horror in this case ― as there was in the Bissonnette one:

It is self-evident that your offending constituted extreme violence. It was brutal and beyond callous — your actions were inhuman. You deliberately killed a threeyear-old infant by shooting him in the head as he clung to the leg of his father. The terror you inflicted in the last few minutes of that small child’s life is but one instance of the pitiless cruelty that you exhibited throughout. There are countless more examples. You showed no mercy. [151]

In Justice Mander’s view,

no minimum period of imprisonment would be sufficient to satisfy the legitimate need to hold you to account for the harm you have done to the community. Nor [would] minimum term of imprisonment would be sufficient to denounce your crimes. [179]

Indeed, in a comment particularly relevant to the comparison the sentence he imposes with Justice Huot’s preoccupation with not letting people spend their lives behind bars, Justice Mander observes that, were he “to impose a minimum period of imprisonment in an endeavour to meet the purposes that I am required to achieve in sentencing you for murdering 51 people, it could not be less than [Mr. Tarrant’s] natural life”. [180] Ultimately, he does

not consider, however long the length of your incarceration during your lifetime, that it could, even in a modest way, atone for what you have done. Ordinarily such an approach would be a poor guarantee of just and proportionate punishment, but I consider yours is one of those exceedingly rare cases which is different. [184]

I think that Justice Mander is right about all of this. Justice Huot, who would no doubt hurl accusations of “sophistry”, populism, and other assorted sins, would not only be wrong but, at the risk of sounding pompous, morally obtuse. Collective indifference and forgetfulness are not just, or even primarily, concerns in relation to those who commit terrible crimes. It would be no less ― and indeed much more ― wrong to be indifferent to the crimes themselves. And it will still be wrong decades from now.

As I recently wrote in discussing an Alberta judgment on the application of section 12 of the Charter, I think that the gross disproportionality test is a sensible construction of its “cruel and unusual punishment” prong, so far as individuals (rather than legal persons) are concerned. Well, I don’t think there is anything grossly disproportional, or indeed disproportional in any way, in denying the possibility of parole to a man who presents himself to a place of worship with the sole purpose of killing as many people as possible, and proceeds to do just that. On the contrary, I think justice may well demand no less. Perhaps there are policy considerations that would explain why a legislature might not put that option on the table. But at the level of principle, I think the New Zealand approach of making the life without parole sentence available in cases where the objectives of punishment cannot be met by a lesser one is right. The Canadian approach of making the parole non-eligibility terms of multiple murderers run consecutively amounts to the same thing, but less transparently, so I think the New Zealand one is preferable.

Granted, the sentencing court should consider repentance and the prospect, even if unlikely on balance, of rehabilitation. There seems to be a difference on this point between the Québec and Christchurch cases, and if this were the reason for Justice Huot’s decision not to impose, in effect, a life sentence without parole, it might have been defensible. (I’m not sure it would have been. Luckily I’m not a judge in charge of sentencing mass murderers, so I get to punt on this question.) But that’s not the main consideration that motivated Justice Huot. On the contrary, he felt strongly enough the need to denounce and punish Mr. Bissonnette that he rewrote (which is a nice way of saying “broke”) the law to impose a 40-year parole ineligibility period, instead of a 25-year one. That suggests that, ultimately, he thought that, as in the Christchurch case, punishment and denunciation dominate. And, if so, a sentence without parole is warranted.

I fully agree with Justices Huot and Mander that the measure of just punishment is not its ability to grab the headlines, and that a civilized justice system must move away from the “an-eye-for-an-eye” instinct. Cases such as these remind us, in any event, the futility of such fantasies. Even if we were in the business of killing murders, we couldn’t kill them six, or fifty-one, times over.

But Justice Mander’s sentencing remarks are a reminder that one need not be vengeful, or to simple-mindedly parrot the tough-on-crime line, to find, in truly shocking and exceptional cases, that the most severe punishment is warranted. Protecting the lives of the citizens is the state’s first responsibility on any plausible view of its role. Providing justice, in the form punishment, in response to those who take their fellow human beings’ lives is the second. In the face of contempt for human life and indifference to, if not actual pleasure in, human suffering, retribution is called for. In extreme cases, locking such people up and throwing away the key is only fair. I do hope that the Québec Court of Appeal, and the Supreme Court if comes to that, take note.

Vavilov in the Prisons

By now, Vavilov—the case in which the Supreme Court re-jigged Canada’s standard of review framework—has received sustained attention, including from yours truly. Over at Administrative Law Matters, Professor Daly has a running post outlining how Vavilov has been applied in some particular interesting cases. And on SSRN, Jamie Chai Yun Liew has an excellent article on the implications of Vavilov for marginalized persons.  Vavilov has had and will continue to have implications for particular areas of law.

My concern in this post is the way Vavilov is being applied, at least in a few cases, in an important area of administrative decision-making: the carceral state. Indeed, it is not often that prisons are mentioned as distinct areas of administrative decision-making. This is, perhaps, because the administrative state is often identified closely with the “welfare state”—a benevolent set of tribunals and decision-makers maximizing benefits for vulnerable people. This is best represented in the deification of labour tribunals as the quintessential administrative decision-maker: expert, oriented towards the public good, efficient, and cheap.

But the prisons are not some separate and apart actor in terms of administrative law. Prison decision-makers operate on delegated power, just as labour and human rights tribunals do. Prisons have become increasingly bureaucratized over time (see Malcolm Feeley and Edward Rubin’s classic text here), and with that bureaucratization has come the potential for the traditional pitfalls associated with administrative decision-making. Indeed, as I will point out in the context of a particular case, “expertise” in the prison context means something very different—it often means expertise in using punitive measures to “manage” “difficult” prisoners.  Coupled with the intersecting disadvantages of prisoners, and the level of control prison administrators—presumably without legal training—hold over those same prisoners, the risk of unlawful or unconstitutional decisions affecting the statutory or constitutional rights of prisoners multiplies. Indeed, prisons were once referred to as “lawless agencies” for good reason—they form, like immigration decision-makers, an intimate part of the administrative state as we know it.

Hence it is important to study the way in which judicial review doctrine works in the carceral state. Indeed, one would expect that Vavilov’s focus on justification might actually make a difference in the prisons, where decisions have historically been made without the close scrutiny of courts. The following two cases illustrate the divergent ways in which, at least in these two cases, Vavilov is faring against the behemothic carceral state. In one of the cases, the court takes a deferential posture based on “expertise,” arguably undermining the justificatory promise of Vavilov. In the other, the court subjects the prison decision to the scrutiny required by Vavilov’s justificatory standard.

Bromby v Warden of William Head Institution, 2020 BCSC 1119

Bromby is serving an indeterminate life sentence for first-degree murder. He was involuntarily transferred from a minimum security institution to a medium security institution in 2019. He objected to this transfer, “[r]aising concerns about the sufficiency of disclosure” in relation to the transfer [2]. Despite this objection, the transfer was approved by the Warden of the minimum security facility. However, Bromby raised a habeas corpus claim, and was successful on that claim. He was transferred back to the minimum security institution.

However, immediately following this decision, the minimum security facility initiated an “emergency involuntary transfer” based on an increase of his security classification score. The final decision to transfer Bromby from minimum to medium security was eventually made by the warden, and was delivered orally to Bromby—but he was not provided written reasons for this decision  in a timely manner, contrary to the Corrections and Conditional Release Regulations [4]. Bromby argued that “the transfer on an emergency basis was unreasonable as he did not pose a threat to the security of the penitentiary or the safety of the inmates or any other person” [5].

In response to this claim, the Warden trotted out an old standard of prison decision-making: the rather specious appeal to “micromanagement” and “expertise.” As the respondent submitted:

  1.    It is not the role of reviewing courts to micromanage prisons even where they feel that intervention measures other than a transfer might have been taken in addressing inmate behaviour. While the applicant may have preferred for other actions to be taken to attempt to manage his risk, deference is owed to the Warden’s assessment that the applicant was unmanageable within Mission (Minimum) Institution.

. . .

  1.    The Decision was . . . reasonable . . . based on the facts and legal constraints before the Warden. The decision of the Warden, a prison administrator familiar with the complexities of Mission (Minimum) Institution and the security classification of inmates, should be provided with deference to decide that the applicant was presently incapable of management within an open perimeter environment. The thorough explanations provided and thoughtful insight as to the specific interventions that the applicant can work towards in becoming a minimum security inmate reflect the Warden’s significant expertise in identifying and managing offender risk. Accordingly, a high degree of deference is owed to the Warden in his decision.


The Court largely accepted this line of thinking:

However, it is the January 2020 Decision that I must assess for reasonableness and determine whether it falls within the range of possible acceptable outcomes which are defensible on the facts and the law. I find that the decision does. That is because the warden is entitled to deference in the decision‑making process. The decision of the warden, a prison administrator familiar with the complexities of Mission Institution and the security classification of inmates, set out the basis for why it was that Mr. Bromby presented as being incapable of management within an open‑perimeter environment [63].

There are two general problems with this line of thinking.

First, I confess that I don’t understand how this line of thinking is at all consistent with Vavilov. Vavilov did away the presumption of expertise for administrative decision-making, instead focusing on “demonstrated expertise” (Vavilov, at para 93). It is true that this demonstrated expertise “may reveal to a reviewing court that an  outcome that might be puzzling or counterintuitive on its face nevertheless accords with the purposes and practical realities of the relevant administrative regime…” (Vavilov, at para 93). This does, fairly, give some latitude for prison decision-makers such as the Warden in this case to apply their knowledge of a particular situation to a dispute. But in this case there is no interrogation of the demonstrated expertise of the Warden. That is, the court does not determine whether the supposed expertise of the Warden was actually demonstrated in the reasons. The fact that, in the court’s view, “[t]he record establishes that the warden turned his mind to all of the relevant considerations…” [65] is not enough to warrant an acceptance of pre-Vavilov law on expertise. This has particular resonance in the prison context, where expertise has often been assumed without demonstration (see Lisa Kerr’s wonderful article here).

I acknowledge that it is genuinely difficult to demonstrate, in the prison context, what actually constitutes “expertise.” Vavilov opens the door to the operational realities of prisons—including issues of security—factoring into a decision. But there is no critical assessment here by the court of how the Warden’s expertise featured into this decision. This seems to be what Vavilov prescribes, and it arguably should factor into any assessment of reasonableness, particularly where the consequences are dire for the individual (Vavilov, at para 133).

Moreover, it is important to recognize that “expertise,” as a general proposition, and especially in the carceral state, can refer to many different things. Simply stating, as the court does in Bromby, that wardens have expertise masks the real question: in what? As Lisa Kerr notes in another outstanding article, at 259, expertise can cut both ways, especially where constitutional rights are at stake. Expertise could be a veneer for stereotyping or discrimination, as I note below.  This is an important normative reason to deny administrative decision-makers a presumption of expertise, which Vavilov explicitly rejects—but which has, based on Bromby, perhaps not filtered down to the carceral state.

Finally, I must acknowledge the old trope about “micromanagement” of prisons, trotted out by the respondent in Bromby. Judicial review is, it is true, not about micromanagement of administrative decision-making. It is about policing the boundaries of statutory schemes designed to cabin administrative activity; it is a traditionally legal and doctrinal activity. As Kerr again aptly acknowledges:

The organization dynamics of prisons tend to resist constitutional constraints, due to the political powerlessness of inmates and the structural isolation of corrections from the community. The status of the inmate is defined in relation to managerial goals, rather than in relation to an externally defined moral norm, and prison managers tend to focus on their vision of scientific management rather than the larger legal order. Amid these institutional tendencies, only the judiciary has the inclination and ability to impose a regular and comprehensive legal framework. The judiciary is a necessary play in prison legality, rather than a necessarily amateur outsider at risk of “micromanagement.” The spirit of habeas corpus, with its strict emphasis on legality and access to justice, so as to challenge deprivations imposed on the physical body, has always had this in mind.

This is fundamentally important. Judicial review is a check against the seductive force of administrative exigency, in which people might be assimilated based on stereotypes or useful organizing tools rather than as individuals. Yet prisons, at the same time, must acknowledge the rights (statutory and constitutional) of prisoners—this is acknowledged in CSC’s enabling legislation.  Ensuring that these rights are upheld is the function of judicial review, which should not be reduced to some afterthought when evaluating the panoply of control mechanisms at the hands of prison decision-makers.

Johnston v Canada (Atorney General), 2020 FC 352

Contrast the previous case with Johnston out of the Federal Court. Johnston involved an inmate at Kent Institution. As a federal inmate, Johnston received payments from Correctional Service Canada “at a modest daily rate” [1]. However, CSC began withholding 100% of Johnston’s modest pay. This is because he had not paid a costs order in favour of the Attorney General. As such, Johnston filed a judicial review of the decision to make the a 100% reduction in his inmate payments.

The Court concluded that, though the relevant statutory scheme gave the CSC the legal authority to make deductions, “it was unreasonable for the CSC to withhold all of the applicant’s income without considering the purpose and principles that govern CSC and without considering the impact the deductions would have on the applicant” [4].

The problem in this case was marred by issues with the record. As the Court noted, the only record of decision was an email chain originating in the CSC. That email chain revealed that there was no “explanation or justification for making deductions from the applicant’s inmate income” apart from the obvious costs order [14]. Specifically, the record did not say “anything about…why the amount of the deduction was set at 100 percent of the applicant’s inmate income” [15].

While Vavilov does note that reasons are not required in every case (Vavilov, at para 77), reflecting well-known law, the case does note that “where reasons are provided but they fail to provide a transparent and intelligible justification…the decision will be unreasonable” (Vavilov, at para 136). This is precisely what happened here. In addition to failing to disclose why the 100% figure was chosen, the CSC failed to consider the vulnerability of the person who had has income reduced. This is directly contrary to Vavilov (see Vavilov, at paras 133 et seq).  The Vavilov majority puts the point eloquently at para 135:

[135]                     Many administrative decision makers are entrusted with an extraordinary degree of power over the lives of ordinary people, including the most vulnerable among us. The corollary to that power is a heightened responsibility on the part of administrative decision makers to ensure that their reasons demonstrate that they have considered the consequences of a decision and that those consequences are justified in light of the facts and law.

The CSC’s conduct in this regard was sorely lacking in terms of the justificatory standard set out in Vavilov—probably more so than Bromby. When a court cannot glean a reasonable justification from the record and reasons—in this case, a generally worded email—the risk increases that the administrative action was arbitrary. The risk increases that, in this particular case, the number of 100% was plucked from the air, without any discernible reason. When we multiply this arbitrariness with the existing vulnerability of prisoners, we have a recipe for administrative disaster.

The point is not that prisoners, because of their vulnerability, must win every judicial review. Doctrine must be applied without fear or favour. But the doctrine includes the consideration of the circumstances of vulnerable persons, and the importance of a decision to those persons, raising the justificatory bar in those cases (see, for this point, Sharif, at para 9). Again, this is not a trump card, but it is an important consideration for administrative decision-makers. It is not something to discard in favour of administrative exigency.

Justification plays a useful role here. It forces the prison, which is by design isolated from the rest of the community, to articulate the reasons for decisions in a way that is understandable to the people within prisons, as well as to the external legal system. Justification is the window by which we can look into the morass of prison decisions, policies, and directives that bear on the actual lives of real people within the carceral state. The tendency in prisons is, likely, to reject these external checks.

That makes those checks all the more important.

Unusual Indeed

The trouble with a caustic, pseudo-originalist opinion of Wakeling JA

In my last post, I described the decision of the Alberta Court of Appeal in R v Hills, 2020 ABCA 263, which upheld a mandatory minimum sentence for the offence of firing a gun into a place “knowing that or being reckless as to whether another person” is there. Two of the judges, Justices O’Ferrall and Wakeling delivered concurring opinions in which they called on the Supreme Court to reconsider its jurisprudence on mandatory minimum sentences and indeed on the interpretation of section 12 of the Canadian Charter of Rights and Freedoms, which protects against “any cruel and unusual treatment or punishment”, more broadly, notably R v Smith, [1987] 1 SCR 1045 and R v Nur, 2015 SCC 15, [2015] 1 SCR 773. I summarized the arguments made by both of the concurring judges in the last post.

Here, I consider specifically Justice Wakeling’s opinion. It is very unusual indeed, in both substance and form. It deploys unorthodox and, in my view, untenable, interpretive techniques, and arrives at startling conclusions. It is long, seemingly scholarly (though there is less real scholarship to it than meets the eye), and caustic. I don’t recall reading anything quite like it in Canada, though admittedly I do not read as many judgments as I would like, especially below the Supreme Court level.

Let me begin with Justice Wakeling’s approach to constitutional interpretation. Justice Wakeling does not explain what he is doing, which is unfortunate, because an explanation might have clarified matters ― not least to Justice Wakeling himself. Be that as it may, what Justice Wakeling seemingly does is resort to a sort of expected applications originalism. This is a way of describing attempts to interpret constitutional provisions by asking how their framers would have expected a question about their application to be resolved. This is a fool’s errand. Serious originalists have long given up on what Benjamin Oliphant and I have described as “speculative transgenerational mind reading”. (126) As Randy Barnett has written, “ascertaining ‘what the framers would have done’ is a counterfactual, not a factual or historical inquiry”. (71)

But Justice Wakeling’s version of expected applications originalism is particularly bad, because he refers to a great extent to events and real or purported beliefs that long predate the enactment of the Charter. Indeed his discussion of the Charter and the views, if any, of its framers is remarkably brief. Justice Wakeling points out that the late

Professor Hogg predicted in 1982 that Canadian courts would give section 12 of the Charter “the same interpretation” courts gave section 2(b) of the Canadian Bill of Rights. In other words, section 12 was of minimal value. Modern political realities made a constitutional death-penalty watch dog unnecessary. And that was the only role section 12 was intended to serve. [217; references omitted]

Most of what Justice Wakeling has to say about the meaning of section 12 goes back much further. The Bill of Rights 1688 is of special significance to him, as he argues that its

prohibition of cruel and unusual punishment was undoubtedly a response, either entirely or, at least partially, to the blood-thirsty sanctions Chief Justice Jeffrey and the other judges imposed on supporters of the 1685 Monmouth Rebellion that challenged the rule of the Catholic King James II during the Bloody Assizes of 1685 and the brutal flogging imposed on Titus Oates for his perjured testimony that cost a large number of Catholics their lives. [148]

From this, Justice Wakeling draws a straight line to the Eighth Amendment to the US Constitution, the Canadian Bill of Rights, and the section 12. He describes the US Supreme Court’s departure from the focus on “horrific penalties” akin to torture and its embrace of disproportionality as a touchstone for assessing violations of the Eighth Amendment in Weems v United States, 217 U.S. 349 (1910) as “judicial heresy”, and writes of the author of the majority opinion in that case that “Justice McKenna’s fingerprints are all over” Smith, [187] and thus subsequent section 12 jurisprudence.

This approach to the interpretation of section 12 makes no sense. Even on an originalist view, why should the meaning of the Charter be determined by what might have been the intentions or expectations not of its framers, but of those of the Bill of Rights 1688, the Eighth Amendment, or even the Canadian Bill of Rights? This isn’t expected applications originalism but expected applications pre-originalism. I know of no precedent or justification for it.

The better originalist approach is that which focuses on the public meaning of constitutional provisions. Historical antecedents are not irrelevant to establishing public meaning (and I have referred to the Magna Carta and the Bill of Rights 1688 myself in writing about section 12 here). However, they are useful in that they ― and their interpretation ― helps us ascertain how a contemporaneous reader would have understood the provision when it was enacted. That being so, the signicance of Weems and subsequent American jurisprudence is very different from that which Justice Wakeling attributes to them. Whether or not they were accurate interpretations of the Eighth Amendment’s original meaning is beside the point. What is noteworthy is that these interpretations would have been part of the context in which section 12 was enacted, and so colour the public meaning the phrase “cruel and unusual punishment” had by the time the Charter was adopted.

A related problem with Justice Wakeling’s approach to interpretation is his use of texts that use wording different from that of section 12 to suggest that the meaning of section must be different. This can be a very useful interpretive tool, but it has to be wielded carefully and honestly. Justice Wakeling relies on three comparisons: with early the constitutions of some American States; with a rejected draft of the Canadian Bill of Rights; and with section 9 of the New Zealand Bill of Rights Act 1990. All of these texts explicitly refer to proportionality, whereas section 12 does not.

Of these, the American texts are somewhat expansive policy statements, of a kind that was mostly ― except, notoriously, in the case of the Second Amendment ― rejected in (what became known as) the US Bill of Rights. The absence of such a statement from the Eighth Amendment doesn’t prove that it disproportionality is not part of its permissible construction. (Somewhat similarly, the absence of an explicit reference to separation of powers, analogous to that found in some State constitutions, in the US Constitution doesn’t mean it does not in fact provide for separate powers.)

With respect to the proposed wording of the Canadian Bill of Rights, Justice Wakeling says that “[a] number of commentators criticized its vagueness”. [201; reference omitted] The concerns of the only such commentator whom Justice Wakeling actually quotes are telling, for he worried, in part about whether a reference to “inhuman or degrading” punishment might be taken to outlaw flogging. Yet Justice Wakeling himself notes that the British “Parliament has repealed the brutal punishments that prompted the 1689 Parliamentary prohibition of cruel and unusual punishments” [153] ― including “flogging”! [154] That commentator’s concerns, in other words, do not deserve to be taken seriously, on Justice Wakeling’s own account. (The reference to flogging is interesting in another way, to which I will shorty turn.) And anyway the exclusion of words like “inhuman” because of their vagueness does not prove that the words retained did not have an element of vagueness calling for construction.

Lastly, the reference to the New Zealand Bill of Rights Act strikes me as quite inappropriate, since that statute was enacted eight years after the Charter. Some of its provisions sought to remedy avoid the Charter‘s real or perceived ambiguities; they tend to be more specific than the Charter‘s. (Compare, for example, New Zealand’s distinct provisions on “freedom of thought, conscience, and religion” and the “manifestation of religion and belief” with section 2(a) of the Charter.) In the case of section 9, one might suppose ― I have not looked into this ― that they New Zealand drafters thought that the outcome of Smith was justified and wrote it into the statute in so many words for the avoidance of doubt. But their choice to do so does not mean tell us anything about the meaning of the Charter, whose drafters were obviously not aware of the subsequent work of their Kiwi counterparts.

The last interpretive issue I will address here is Justice Wakeling’s reading of section 12 as a mere enumeration, and a remarkably brief one at that, of prohibited punishments. One striking consequence of this reading is that Justice Wakeling thinks that, because imprisonment was a commonly used punishment when the Charter was enacted and thus not unusual,

section 12 … does not allow a court to declare jail sentences cruel or unusual punishments. … [O]ffenders may not invoke section 12 to challenge either mandatory-minimum or mandatory-maximum jail sentences or any other jail sentence. [244]

(It is worth noting that Justice O’Ferrall “question[s]” [115] and indeed seems to reject this view.)

Justice Wakeling repeats a mistake committed by Justice Scalia, including in his comments on the Eighth Amendment in the famous lecture “Originalism: The Lesser Evil”. Justice Wakeling refers to some of Justice Scalia’s decisions seeking to limit the import of the Eighth Amendment to the 1791 catalogue of barbarity ― but not to that lecture where, tellingly, Justice Scalia professed being a “faint-hearted” originalist, because he wouldn’t bring himself to countenance the punishment of flogging even if was practised in 1791. The catalogue approach, it seems, doesn’t really work.

In a lecture of his own, “Scalia’s Infidelity: A Critique of Faint-Hearted Originalism“, Randy Barnett explains why. He points out that

original public meaning originalism attempts to identify the level of generality in which the Constitution is objectively expressed. Does the text ban particular punishments of which they were aware, or does it ban all cruel and unusual punishments? (23)

As Professor Barnett notes, “[t]his is not to say … the broader provisions of the text lack all historical meaning and are open to anything we may wish them to mean”. (23) But that meaning, if there is one, must be established with reference to the time of those provisions’ enactment ― not to a period that preceded it by two or three centuries. Justice Wakeling’s own reasons suggest that, whatever may have been the case in 1689 or even 1791, the phrase “cruel and unusual” may well have acquired a broad and morally loaded meaning by 1982. He does not even contemplate this possibility.

This leads me to concerns about the form and tone of Justice Wakeling’s opinion. It has an air of scholarliness: at over 12,000 words and 200 footnotes, it has the heft of an academic article. And yet this is only an appearance. It is inimical to good scholarship ― even, I would argue, in a judicial opinion, and not only in an academic setting ― to ignore counter-arguments and relevant sources that do not support one’s claims. Meanwhile, a great many of those footnotes turns out to cite to Justice Wakeling’s own opinions; a flaw of much academic writing, my own not excepted, but manifested here to an inordinate degree.

And then there is the bitter vehemence of Justice Wakeling’s writing. From the outset, he heaps scorn on the Supreme Court’s precedents, calling the “reasonable hypothetical” approach to section 12 they command “remarkable, to say the least”, [124] and claim that “[t]here is no constitutional doctrine that justifies this unusual method”. The decision in Smith is “surprising[]” [219] and “unexpected”. [220] “The contribution” that an argument made by Justice Lamer ― to whom Justice Wakeling denies the courtesy of a “as he then was” ― “makes to the debate is difficult to comprehend”. [226]

But Canadian courts and judges are not the only targets of Justice Wakeling’s contempt. I have already referred to his desription of Weems as “heresy”. If this were said about a fellow judge on Justice Wakeling’s court, this would be as mean as any of Justice Scalia’s cantankerous dissents. Still, such disagreements can appropriately be aired. But judges do not normally take it upon themselves to critique their colleagues in other jurisdictions. Not only is Justice Wakeling not qualified to pronounce on what it orthodox and what is heretical under American law ― it’s just not his job. Not content with commentary on the past, however, Justice Wakeling dabbles in political prognostication too, declaring that he

suspect[s] that the likelihood that additional states will abolish the death penalty is probably about the same as the likelihood that the Supreme Court – with a majority of conservative-minded justices – will sanction additional limits on the availability of the death penalty. [181; reference omitted]

To be clear, I have no objection to a judge expressing disagreement with the jurisprudence of a higher court. On the contrary, judicial criticism of binding authority ― so long as that authority is followed ― can be valuable; no less, and arguably more, than that of scholars and other commentators. If the lower courts are saying that a legal doctrine is not working well, the higher courts would do well to listen ― though they need not agree, and they should not agree in this case, as I argued in my last post. Justice O’Ferrall’s opinion strikes me as perfectly fine. But not so Justice Wakeling’s.

I have been tone-policed enough to be wary of engaging in such critiques myself. But Justice Wakeling is, after all, a judge ― and I think that judges can rightly be held to a standard of equanimity that should not be applied to academics, whose role it is to critique, and sometimes criticize, the exercise of the judicial power. I have also defended the use of strong language in judicial opinions. Still, there are lines not to be crossed. A judge ought not to be dismissive or petulant; nor should he engage in political commentary or, I think, make any sort of pronouncement on the merits of the laws (enacted or judicially articulated) of other countries. Justice Wakeling is and does all of these things. If he wants to act like a politically preoccupied professor, he should resign his judicial office.

I do not know how widespread the views expressed by Justices O’Ferrall and Wakeling are. Perhaps the Supreme Court will take heed and reconsider its jurisprudence in relation to section 12. In any case it will face other difficult questions about the interpretation of the Charter. Justice Wakeling’s opinion illustrates a number of things not to do in such cases. Courts should not look to the ways the authors of constitutional provisions, let alone the authors of their predecessors, expected these provisions to be applied. They should not be careless, let alone deliberately unfair, when they compare different texts. They should not convert moral language into laundry lists. And, of course, they should not be mean-spirited. Justice Wakeling’s opinion is unusual indeed, and I hope it stays that way.


Judges of the Alberta Court of Appeal question the Supreme Court’s jurisprudence on mandatory minimum sentences

Last month the Alberta Court of Appeal issued an interesting decision that concerned the constitutionality of yet another mandatory minimum sentence, this one in section 244.2 of the Criminal Code, for “intentionally discharg[ing] a firearm into or at a place, knowing that or being reckless as to whether another person is present in the place”. The mandatory minimum is four years’ imprisonment (or more if organized crime is involved). For fairly straightforward rasons given by Justice Antonio, R v Hills, 2020 ABCA 263, upholds the four-year mandatory minimum, rejecting the claim that it is “cruel and unusual” within the meaning of section 12 of the Canadian Charter of Rights and Freedoms.

But in separate concurring reasons Justices O’Ferrall and Wakeling go on to criticize the Supreme Court’s approach for dealing with such cases. Both concurring opinions raise important questions, not only about the correct approach to mandatory minimum sentences under section 12 of the Charter, but also about constitutional interpretation and construction more broadly.

In this post, I summarize Justice Antonio’s lead opinion, as well as the common aspects of the two concurring ones, and explain why I think the Supreme Court is right and Justices O’Ferrall and Wakeling are wrong about section 12. In a follow up post, I shall write in some detail about Justice Wakeling’s opinion, which is startling, and startlingly wrong, in its method and tone, and deserves special attention and criticism.

In R v Smith, [1987] 1 SCR 1045 and, more recently, R v Nur, 2015 SCC 15, [2015] 1 SCR 773, the Supreme Court held that a mandatory minimum sentence that is “grossly disproportionate” to the gravity and blameworthiness of an offence is “cruel and unusual” within the meaning of section 12. Gross disproportionality can be shown either in the particular case or, alternatively ― and controversially ―, in a reasonable hypothetical, a set of circumstances that can be expected to occur and that would be captured by the impugned provision. This is the approach that the accused in Hills took.

Mr. Hills pleaded guilty to having repeatedly fired a rifle “suitable for big game hunting” [4] into the walls and windows of a family residence ― among with less frightening misdeeds, all part of a rampage undertaken in a state that a former Toronto mayor would have described as drunken stupor. The sentencing judge considered that the mandatory minimum would not be grossly disproportionate to his offence, but it could be in a reasonable hypothetical, mainly because the applicable definition of “firearm” captures weapons shown by an expert to be incapable of penetrating a typical building wall. One could therefore reasonably imagine the four-year sentence being imposed on a person who fired a weapon “at a place” whose occupants were not thereby endangered. The judge sentenced Mr. Hills to three and a half years’ imprisonment.

Justice Antonio (with whose reasons Justice O’Ferrall agrees, so far as they go) considers this to be an error. This is because the shots fired even from low-power weapons might “penetrate a door or window”. [80] Moreover, the weapons or the shots might alarm bystanders or the people inside the place at which they are fired, and generally undermine “the feeling the safety in communities”. [82] Justice Antonio also refers to Nur, where

 a 40-month sentence was imposed on a 19-year-old first offender who merely possessed a loaded firearm in a public place for a short period of time and did not discharge it or use it in a threatening manner. If 40 months was an appropriate sentence in the Nur case, then an additional eight months as a mandatory minimum penalty where a firearm was actually used does not amount to a grossly disproportionate sentence.

Justice Antonio concludes that a fit sentence for Mr. Hills would be four and a half years’ imprisonment.

As mentioned above, Justices O’Ferrall and Wakeling both call for the Supreme Court’s decisions in Smith and Nur to be revisited insofar as they require the courts to undertake gross disproportionality analysis based on reasonable hypotheticals, and not only the facts before the sentencing court. Some of the arguments they make are similar. I address them here. Justice Wakeling’s opinion also makes additional points not raised by Justice O’Ferrall. I turn to them below.

The main argument on which Justices O’Ferrall and Wakeling rely is that the use of reasonable hypotheticals to test the constitutionality of mandatory minimum sentences is inconsistent with the import of section 12. Justice O’Ferrall argues that

[a]n interpretation [of the Charter] which relies on the presumed detriment to a non-existent offender if a certain term of imprisonment is imposed is not an interpretation which a citizen would contemplate. It is an interpretation which might legitimately surprise the citizen. It does not flow logically from the text of s.12 of the Charter. [108; see aslo Justice Wakeling’s comment at [126]]

For Justices O’Ferrall and Wakeling, since section 12 protects an individual “right not to be subjected to any cruel and unusual treatment or punishment”, only the situation of the offender before the court can be taken into consideration, and the courts should avoid invalidating provisions that might only hypothetically result in unconstitutional applications. Just as laws are not invalidated because they might be invoked to effect unconstitutional arrests, they should not be disturbed because they might, in some cases, lead to unconstitutional sentences. As Justice O’Ferrall puts it, “[b]ut for the approved reasonable hypothetical analysis, the accused could [sic] care less about the constitutionality of the law. His complaint is with respect to his treatment or punishment”. [109]

Indeed, Justices O’Ferrall and Wakeling reject the test of “gross disproportionality” itself, which the Supreme Court has long used as a proxy for deciding whether a punishment is cruel and unusual. Justice O’Ferrall argues that

A sentence may be disproportionate from the perspective of both the offender and the offence and yet … prescribed to achieve the fundamental purpose of sentencing, namely protecting society. Even a grossly disproportionate sentence may not be found to constitute cruel and unusual punishment if, for example, in order to stem the tide of a deadly pandemic, Parliament found it necessary to prescribe extremely harsh punishments for what otherwise might be regarded as minor misdemeanors. [117; see also Justice Wakeling’s comment at [132]]

I do not think that any of this is right.

Start with the meaning of section 12. The concurring opinions go wrong because they fail to distinguish between the interpretation and the construction of constitutional provisions. Interpretation is the activity of ascertaining the communicative content of the text. Construction is the elaboration of doctrines that allow the text to be given legal effect. Some cases can be resolved at the interpretation stage. As I have argued here, the interpretation of section 12, and specifically of the word “cruel”, can tell us that this provision does not protect corporations. But in other cases courts need to engage in (good faith) construction to apply vague language ― and that of section 12 is vague, if not quite as vague as some commentators would have believe.

The word “cruel” is not infinitely malleable, but it is not self-explanatory either. Unless they are going to rely on seat-of-the-pants impressionistic decision-making in every case, courts need to work out a consistent way to determine whether a given sentence is cruel and unusual. This is an exercise in construction, which is a form of legal reasoning. Unlike in the realm of interpretation, the presumed (actually, purely conjectured) reactions of reactions of citizens are not a useful guide to what the courts should do here. The courts’ task is not to avoid surprises ― the framers of the constitution make a certain degree of judicial creativity unavoidable when they use vague language ― but rather, as Randy Barnett and Evan Bernick have argued, to give effect to the purpose of the provision.

Is the test of gross disproportionality a misguided construction of section 12? In my previous post on that provision’s meaning (linked to above) I have suggested that it is not, so far as the punishment of natural persons is concerned. I wrote that “disproportionality can be a useful indication of cruelty”, provided that “also causes or reflects indifference to suffering”, which may “always be the case with grossly disproportional punishment is inflicted on human beings”. Justice O’Ferrall’s example is ambiguous and does not persuade me. It may be taken to suggest that in the circumstances of “a deadly pandemic” “what otherwise might be regarded as minor misdemeanors” become extremely blameworthy crimes. If so, there is no gross disproportionality in punishing them harshly, so long as the relevant circumstances exist. But if Justice O’Ferrall suggests that a public emergency justifies harsh punishment of unrelated offences, I don’t see how that follows.

If not the gross disproportionality test, is the reasonable hypothetical approach an impermissible construction of section 12? Actually, I think there are very good reasons for the courts to adopt it. Contrary to what Justices O’Ferrall and Wakeling say, a mandatory minimum sentence impacts an offender as to whom it would not be cruel and unusual, albeit indirectly. As Justice Arbour explained in her concurrence in R v Morrisey, [2000] 2 SCR 90,

mandatory minimum sentences … must act as an inflationary floor, setting a new minimum punishment applicable to the so-called ‘best’ offender whose conduct is caught by these provisions.  The mandatory minimum must not become the standard sentence imposed on all but the very worst offender who has  committed the offence in the very worst circumstances.  The latter approach would not only defeat the intention of Parliament in enacting this particular legislation, but also offend against the general principles of sentencing designed to promote a just and fair sentencing regime and thereby advance the purposes of imposing criminal sanctions. [75] 

Justice Wakeling’s own reasons illustrate this dynamic. He breaks down the range of sentences permitted by Parliament into bands for the least and most serious cases, and those in the middle. On this approach, if Parliament enacts or raises the mandatory minimum, the sentences of most offenders, except perhaps the very worst ones, go up. Of course, Parliament is entitled to intervene in sentencing. But the fact that its intervention impacts all offenders means that it is appropriate to consider its constitutionality even in cases where the minimum sentence would not have been cruel and unusual. At the risk of mixing metaphors, I think it’s not an implausible construction of section 12 to say that it does not permit the inflationary floor to be sullied by the cruelty of sentences required to be imposed even on some, albeit not all, offenders.

The other reason for the courts to continue to police reasonable hypotheticals might sound more in policy, but it too is relevant to section 12. It is plea bargaining. A prosecutor can threaten an accused person with a high mandatory minimum sentence so as to secure a guilty plea to some other, less serious offence. By the very nature of such situations, there is no scope for the mandatory minimum to be challenged; indeed the offence to which it is attached never even features before a court. But to the extent that the mandatory minimum has served to secure a guilty plea from a person who might be innocent (or at least might be able to raise a reasonable doubt about his or her guilt), its deployment by the prosecutor is, arguably, a form of cruel and unusual treatment that offends the Charter.

It has been set that the judicial response to the last Conservative government’s “tough on crime” agenda has been nothing less than a rebellion. Justice Wakeling professes himself “extremely troubled by the fact that Canadian courts have been busy striking down Criminal Code provisions that impose mandatory-minimum sentences”. [123] The concurring opinions in Hills are a counter-rebellion of sorts, directed not against Parliament but against the Supreme Court.

But the rebels are wrong. Their approach to constitutional text, which collapses interpretation and construction and oversimplifies constitutional meaning is not compelling. They fail to see the repercussions of mandatory minimum laws that deserve the suspicion with which the courts have treated them. The Supreme Court has often read constitutional provisions ― both power-conferring and rights-protecting ones ― more expansively than it should have. But I am not convinced that this is the case with section 12 of the Charter.

PS: I have neglected blogging on judicial decisions in the last couple of months, and will try to make up at least some of this backlog. If you have a case I should get onto in mind, please do get in touch.

La Constitutionnalité de l’application de la Loi 101 aux entreprises fédérales

Le 18 août dernier, le ministre responsable de la Langue française, M. Simon Jolin-Barette, a annoncé qu’il souhaitait voir la Charte de la langue française appliquée aux entreprises sous juridiction fédérale. Une telle mesure forcerait notamment les entreprises fédérales à obtenir un certificat de francisation et à se soumettre à une série d’obligations destinées, comme l’indique le préambule de la Charte, à faire du français « la langue normale et habituelle du travail, […] des communications, du commerce et des affaires ». On estime qu’au Québec, 135 000 travailleurs du secteur privé ne sont ainsi couverts ni par la Charte de la langue française, ni par la Loi sur les langues officielles. Or, cette proposition soulève de sérieux doutes quant à sa constitutionalité, notamment quant à savoir si l’Assemblée nationale a le pouvoir législatif nécessaire pour procéder seule à cet amendement. 

L’Assemblée nationale peut-elle procéder seule?

Tout d’abord, il ne fait aucun doute que les « institutions fédérales » comme les départements gouvernementaux ou les sociétés d’état fédérales qui, elles, sont déjà encadrées par la Loi sur les langues officielles, ne sauraient en aucun cas être visées par la législation provinciale. Le chapitre V de la Loi sur les langues officielles établit un régime juridique qui garantit le droit de travailler dans l’une ou l’autre des deux langues officielles. Il s’agit là d’un cas clair de prépondérance fédérale, en vertu de laquelle une loi fédérale valide rend inopérante une loi provinciale autrement valide avec laquelle elle est en conflit. 

En ce qui a trait aux entreprises sous juridiction fédérale comme les banques, les entreprises ferroviaires, maritimes, de transport interprovincial et de télécommunications, il est fort probable que les tribunaux jugent que leur assujettissement à des mesures réglementaires linguistiques excède la juridiction de la province. Tel qu’indiqué dans Devine c Québec (procureur général), [1988] 2 RCS 790, la juridiction sur la langue, qui n’est pas explicitement prévue au partage des compétences de 1867, doit être rattachée à un champ de compétence. La langue de travail est considérée par la jurisprudence comme relevant des relations de travail. Ainsi, dans les dernières décennies, plusieurs décisions ont confirmé la compétence fédérale en matière linguistique au sein des entreprises fédérales ainsi que l’inapplicabilité de la Charte de la langue française aux entreprises fédérales situées au Québec, dont Joyal c Air Canada, 1976 QCCS 1211 à la p 1230 et Association des Gens de l’Air du Québec Inc. c Lang, [1977] 2 CF 22 au para 39, ainsi que plus récemment Girard c Telus Québec inc., 2006 QCCRT 236 et Léveillé c Conseil canadien des Teamsters, 2011 CCRI 616.

Or, des auteurs ont récemment affirmé que des revirements jurisprudentiels en matière d’exclusivité des compétences justifiaient un changement de paradigme sur cette question. Se basant sur l’arrêt Banque canadienne de l’Ouest c. Alberta, 2007 CSC 22, ils affirment que, du fait que la doctrine d’exclusivité des compétences ne s’applique que lorsqu’une mesure législative entrave le contenu « essentiel » de la compétence de l’autre ordre de gouvernement, la Charte de la langue française peut être appliquée aux entreprises fédérales. Pourtant, ce raisonnement ne tient pas la route et ce, pour deux raisons. D’abord, il fait abstraction de la jurisprudence pertinente en la matière. Ensuite, il sous-estime l’impact qu’a la Charte de la langue française sur les activités habituelles d’une entreprise.

Premièrement, pour conclure que la Charte de la langue française peut s’appliquer à des entreprises fédérales, ces auteurs écartent une importante décision, NIL/TU,O Child and Family Services Society c B.C. Government and Service Employees’ Union, 2010 CSC 45. Dans cette affaire, la Cour suprême tranche que le test de l’exclusivité des compétences n’est pas approprié pour examiner les questions de compétences en matière de relations de travail, lui préférant un test en deux étapes. La première étape est le test du « critère fonctionnel ». Il faut alors examiner la nature de l’entité, son exploitation et ses activités habituelles pour voir s’il s’agit d’une entreprise fédérale. Dans un tel cas de figure, les relations de travail seront assujetties à la réglementation fédérale plutôt qu’à la réglementation provinciale. Si – et seulement si – la première étape du test n’est pas concluante, il faut alors se tourner vers la seconde et se demander si la mesure proposée entrave le cœur de la compétence fédérale. 

En l’espèce, il ne fait aucun doute que l’application du test du « critère fonctionnel » mine toutes les chances de Québec de voir la Charte de la langue française être appliquée aux entreprises fédérales. Pas question ici de se demander si la Charte de la langue française entrave le contenu « essentiel » du chef de compétence fédéral. Le simple fait que les activités habituelles des entreprises fédérales soient justement de nature fédérale suffit à les soustraire à l’application de la Charte de la langue française.

Deuxièmement, même si c’était le test de la doctrine de l’exclusivité des compétences qui devait être retenu, il est clair que la Charte de la langue française entrave le « contenu essentiel » des champs de compétence fédéraux correspondants, comme la poste, les banques, le transport interprovincial, la navigation, etc. Le critère de l’entrave n’équivaut pas à une paralysie ou une stérilisation selon Rogers Communications Inc. c Châteauguay (Ville), 2016 CSC 23au para 70. Ainsi, dans Banque de Montréal c Marcotte, 2014 CSC 55 au para 66, la Cour suprême du Canada laisse entendre qu’une loi provinciale sera déclarée invalide si elle «restreint» les activités d’une entreprise fédérale. Or, la Charte de la langue française est on ne peut plus intrusive. Elle exige des entreprises qu’elles se soumettent à une analyse étendue de leurs activités (art 141). Elle régit notamment la langue de communication d’un employeur avec ses employés (art 41), des offres d’embauche et de promotion (art 41), des conventions collectives (art 43), interdit de congédier, de mettre à pied, de rétrograder ou de déplacer un employé qui ne parle pas assez bien une langue autre que le français (art 45), interdit d’exiger à l’emploi une langue autre que le français si ce n’est pas nécessaire (art 46), etc. De plus, le fait pour une entreprise de ne pas se plier aux exigences de la Charte de la langue française peut être lourd de conséquences. La politique gouvernementale en matière linguistique prévoit que les entreprises de 50 employés et plus qui ne possèdent pas de certificats de francisation ne se verront accorder ni contrat, ni subvention, ni avantage par l’administration publique. De plus, les amendes prévues à la Charte de la langue française pour une première infraction peuvent aller jusqu’à 6 000$ pour un particulier et jusqu’à 20 000$ pour une entreprise (art 205).

Existe-t-il des alternatives?

Ainsi donc, si la réglementation linguistique des entreprises fédérales relève du Parlement, comment le gouvernement québécois pourrait-il s’y prendre pour faire appliquer la Charte de la langue française aux entreprises fédérales? 

Premièrement, certains auteurs ont suggéré que le Parlement pourrait déléguer aux provinces son pouvoir législatif en matière linguistique. Toujours selon ce courant de pensée, le gouvernement du Québec pourrait demander au Parlement de lui déléguer son pouvoir de réglementer l’utilisation de la langue française au sein des entreprises fédérales. Or, il semble que ce raisonnement soit erroné. En effet, s’il est vrai, comme le font remarquer ces auteurs, que le Parlement peut légitimement déléguer ses pouvoirs réglementaires linguistiques à un territoire comme il l’a fait pour le Nunavut, la délégation aux provinces de pouvoirs législatifs est proscrite par les tribunaux depuis Nova Scotia Inter-delegation, [1951] SCR 31

Deuxièmement, Québec pourrait demander au Parlement qu’il incorpore à sa propre législation un renvoi à la Charte de la langue française. Cette façon de procéder a été reconnue comme étant valide par les tribunaux canadiens depuis Coughlin v Ont. Highway Transport Bd., [1968] SCR 569. En vertu de cette méthode de référencement, toute modification ultérieure de la Charte de la langue française par l’Assemblée nationale s’appliquerait immédiatement et automatiquement aux entreprises fédérales en vertu de A.G. for Ontario v Scott, [1956] SCR 137. Il s’agit là d’une façon de procéder plus respectueuse des principes de droit constitutionnel canadien, mais aussi de celle qui requiert le plus de volonté politique. La Charte de la langue française a toujours fait l’objet d’un feu nourri de critiques au Canada anglais et il serait plutôt surprenant de voir le gouvernement fédéral l’adopter implicitement en y faisant référence dans sa propre législation. D’ailleurs, si le Parti Conservateur, le NPD, le Bloc Québécois et le Parti Vert s’étaient engagés aux dernières élections à faire appliquer la Charte de la langue française aux entreprises fédérales conformément à la demande du premier ministre du Québec, le Parti Libéral, lui, n’avait pas fait de même. Justin Trudeau s’était d’ailleurs opposé par le passé à un renforcement de la Charte. Parions que des discussions musclées sont à venir dans les prochains mois entre Ottawa et Québec.