Conventional Thinking

There is big news on the Supreme Court appointment front today, which is arguably not getting enough attention. According to the Globe’s Sean Fine, “[t]he Conservative government has turned to Quebec to create a candidate list for the Supreme Court of Canada” ― asking the provincial government to submit names of potential replacements for Justice Fish (who retired last fall) and Justice Nadon (whose appointment the Supreme Court invalidated in March). The federal government is promising to choose the new judge from the provincial list. Relying on a federal government “source”, Mr. Fine writes that

the new process is not meant to be a precedent … It applies to the current vacancy, but will probably not be used to select a replacement for Justice Louis LeBel of Quebec when his retirement takes effect at the end of November. It would be wrong to see the same process being used for the two judges, the federal source said.

Except that Québec’s Justice Minister does not see it that way. In her view, “the collaboration undertaken with [her] federal counterpart will allow us to chart the course for things to come.”

What we are witnessing, at least for this appointment, but perhaps for the future too, if the Québec government has its way (and perhaps those of other provinces, which would not want to miss out on such an expansion of their powers), is nothing less than the implementation of the appointment process that would have been constitutionalized (in a proposed section 101C of the Constitution Act, 1867) by the (failed!) Meech Lake Accord. In the mid-1980s it was thought that this sort of change required a constitutional amendment, debated over years of public agony; in the mid-2010s, it can apparently be done by some phone calls, about which we learn weeks after the fact and might not care. But what about those big huge Supreme Court decisions this spring, in the Nadon case and in Reference re Senate Reform, 2014 SCC 32? didn’t those have something to say about changing the constitution, in particular as it concerns the Supreme Court? Well, they did. But the events might be exposing the limits of the Court’s pronouncements, and indeed of its power, faster than anyone probably expected.

In the the Nadon decision, the Supreme Court’s majority was of the opinion that

[u]nder s. 41(d) [of the Constitution Act, 1982], the unanimous consent of Parliament and all provincial legislatures is required for amendments to the Constitution relating to the “composition of the Supreme Court”. The notion of “composition” refers to ss. 4(1), 5 and 6 of the Supreme Court Act, which codify the composition of and eligibility requirements for appointment to the Supreme Court of Canada as they existed in 1982 (par. 91).

It further found that all other “essential features of the Court” (par. 94) were entrenched under par. 42(1)(d) of the Constitution Act, 1982, thus being subject to amendment under the “7/50” formula. Interestingly, the majority’s opinion omits the Supreme Court Act‘s reference to the appointment process ― subs. 4(2), which provides that “[t]e judges shall be appointed by the Governor in Council by letters patent under the Great Seal” from the list of provisions entrenched by s. 41. Nor is it entirely clear whether it is entrenched by par. 42(1)(d). The majority’s opinion states that “the essential features of the Court” which are so entrenched are to be

understood in light of the role that it had come to play in the Canadian constitutional structure by the time of patriation. These essential features include, at the very least, the Court’s jurisdiction as the final general court of appeal for Canada, including in matters of constitutional interpretation, and its independence.

The appointment mechanism is conspicuously absent from this meagre list, but then the list is not exhaustive. It seems at least logical to suppose that it is, in fact, an “essential feature” of the Court. And, to reiterate, in the years following patriation, the political actors presumably thought that it was entrenched and required constitutional amendment to be changed.

Now, subs. 4(2) of the Supreme Court Act only mentions appointment by the Governor in Council. It doesn’t say anything about any procedure that must, may, or may not be followed prior to that appointment. Sounds familiar? It should. That’s also the situation with respect to appointments to the Senate under s. 24 of the Constitution Act, 1867, which provides that “The Governor General shall from Time to Time … summon qualified Persons to the Senate.” In the Senate Reform Reference, the government argued that this silence about what preceded appointments allowed it to organize “consultative” elections the winners of which the government would have to “consider” recommending to the Governor General for appointment. The Supreme Court rejected this claim, first and foremost on the basis that “consultative” elections “would fundamentally alter the architecture of the constitution” (par. 53).

Unfortunately, the Court does not define this notion of constitutional “architecture” well at all, beyond saying that it has something to do with “[t]he assumptions that underlie the text and the manner in which the constitutional provisions are intended to interact with one another” (par. 26). The Senate’s place in the constitutional “architecture” is that of a chamber of “sober second thought,” devoid of democratic legitimacy and the ability to challenge the House of Commons that elections would confer on it. But what about the Supreme Court?

In my view, there is a strong argument to be made that it was always an “assumption[] underlying the text” both of s. 101 of the Constitution Act, 1867, which authorized Parliament to create the Supreme Court in the first place, and then of the amending formula of the Constitution Act, 1982, that the federal government would be in control of the appointments to the Supreme Court, and that provincial governments would not be part of it. Again, it is based on this assumption that the inclusion of the provinces in the process was stipulated in the Meech Lake Accord for constitutional amendment and not just implemented on an informal basis.

So is the involvement of a provincial government in the appointment of a Supreme Court judge unconstitutional in the same way as “consultative elections” to the Senate? Perhaps not. In a paper dealing with the federal government’s Senate reform plans (published before the Supreme Court heard, much less decided the Senate Reform Reference), Fabien Gélinas and I have argued that there is a distinction to be made between a law setting up “consultative elections” and an informal process of genuine consultation prior to the appointment of a Senator:

the Prime Minister is free to consult before advising the Governor General to summon a person to the Senate. However …  when that consultative process is made the subject of legislation, that legislation—unlike a mere practice—must pass constitutional muster. The distinction between practice and legislation is … what allows constitutional conventions to grow even when the conventional rule is at odds with a clear rule of constitutional law. The legal limits on the constitutional changes that Parliament can achieve by legislating are more stringent than the political limits that constrain the actual action of the Prime Minister. This is only logical, because a law, once enacted by the Parliament of today, needs no further confirmation by those of the future to remain in force, whereas the practice of a single Prime Minister will not acquire the binding character of a convention unless his or her successors come to view it as “the constitutional position” and feel bound by it themselves (p. 122; reference omitted).

The involvement of the Québec government in the forthcoming Supreme Court appointment is informal; even if the process is repeated when Justice Lebel is replaced, it will still be a mere practice, not the subject of legislation. The Québec government will have to hope that the current Prime Minister’s successors will feel bound to replicate the practice for it to crystallize into a firm constitutional convention.

Yet the notion of a “constitutional architecture” challenges the orthodox distinction between law and convention on which we relied (without quite believing ― certainly on my part ― in the distinction’s validity). It may be that some of the current constitutional conventions are part of the entrenched “architecture.” It may also be that the “architecture” prevents the development of new conventions that would undermine it. I don’t think that anyone ― including the members, current or future, of the Supreme Court ― know whether it does. And it is doubtful whether we will find out. If both the federal and the provincial government are on board, no one will begin a reference that would force the Supreme Court to clarify its position. Could another Rocco Galati challenge the appointment that will come out of the current process? Perhaps, but I wouldn’t count on it happening. One thing that seems reasonably certain, however, is that although conventional thinking may no longer hold as a result of this spring’s decisions by the Supreme Court, thinking about constitutional conventions, which the Court avoided, will still be necessary.

UPDATE: I explain why, regardless of its constitutionality, making permanent this change in the appointments process is a bad idea in this post.

Constitutional Amendment and the Law

I have been a bit harsh on the Supreme Court in my first post on its opinion in the Reference re Senate Reform, 2014 SCC 32, saying that it had reduced the constitutional text to the status of a façade, which hid as much as it revealed of the real constitutional architecture, which only the Court itself could see. But one must recognize that the Court’s position was very difficult.  The amending formulae codified in Part V of the Constitution Act, 1982, are a nightmare, at once too precise and too vague to guide their interpretation. Although in our legal system text, especially constitutional text, is supposed to be the legal form par excellence, superior to any unwritten norm, Part V shows that this is not always so.

It is often said that, before Part V was added to the constitution in 1982, there was no general amending formula in the Canadian constitution. That is only true if “constitution” is understood as “constitutional text.” In reality, there was an amending formula ― the Canadian constitution could be amended by the Imperial (i.e. British) Parliament, which in accordance with a “constitutional position” (i.e. convention) recognized by the Preamble of the Statute of Westminster, 1931, would only act on address of the Canadian Parliament, which, in accordance with a further convention of which the Supreme Court recognized the existence in the Patriation Reference, could only make such an address with “substantial provincial consent.”

This last convention, requiring substantial provincial consent to constitutional changes, was obviously somewhat vague. And indeed it often said that vagueness is an inherent limitation of constitutional conventions, and perhaps one of the reasons which prevent conventions from attaining legal status. More generally, in his great work on The Concept of Law, H.L.A. Hart argued that the passage from somewhat uncertain traditional rules to formal ones was part of a movement from a pre-legal to a legal system. The replacement of the convention requiring “substantial provincial consent” with specific, written amending formulae forming part of the constitutional text ought to have clarified the constitutional rules, and made them more law-like.

Instead, what we got is a system which is in many ways no clearer than the old conventional rule. Indeed, Part V illustrates Lon Fuller’s insight that an ostensibly legal rule or system of rules can fail certain formal requirements (of what he called the “inner morality of law” and what we usually refer to as the Rule of Law) to the point where they fail to guide behaviour and, thus, to be law at all.

The system of a general rule (s. 38 of the Constitution Act, 1982), examples of the general rule (s. 42), and exceptions to the general rule (ss. 41, 43, 44, and 45, some of which (ss. 44 and 45) themselves sound like plausible general rules) does not make for consistency, which is one of the Rule of Law requirements outlined by Fuller. (I note, however, that this system is somehow very Canadian, in that it parallels that which we have adopted for dividing powers between Parliament and the provinces: there, the “peace, order and good government” clause of s. 91 of the Constitution Act, 1867 is the general rule, followed by examples of federal powers in s. 91, and exceptions in s. 92, at least one of which, subs. 92(13) was itself very broad. Not coincidentally, this complex scheme arguably contributed to the distribution of powers being interpreted in a way that is probably far from what its authors had intended.) The mention of the Supreme Court in the amending formula ― combined with the conspicuous absence of the Supreme Court Act from the list of enactments composing the “constitution of Canada” is another glaring example of the inconsistency of Part V.

What is more, its rules are not exemplars of clarity (does, for instance, the “selection of Senators” refer only to their formal selection by the Governor General, as the federal government argued, or to the whole process leading to it?) Some of these rules also seem to produce results so absurd as to border on the impossible (for instance, as one of the judges suggested at the hearing of the Senate Reference, the amending formula seems to indicate that Canada could be turned into a dictatorship more easily than into a democratic republic).

Add this all up, and we have a set of amending formulae that, as Fuller predicted, fail to guide behaviour ― not only that of the politicians to whom they are addressed in the first instance, but also of the courts to which the politicians turn for help understanding them. We have, in other words, a set of rules which, although purportedly legal, indeed purportedly part of the “higher law,” in some circumstances fail to be law at all. (One should not exaggerate the scope of the problem. In many cases ― say, transforming Canada into a republic ― the import of Part V will be perfectly clear. But the Senate Reference as well as l’Affaire Nadon show the importance of cases where this is not so.)

Yet if one thing is unmistakable after the entrenchment of Part V, it is that the “procedure for amending the constitution of Canada” is a legal, and no longer a conventional matter. The courts are stuck with it, and cannot offload the problem of interpreting it to politicians. (In reality, the Supreme Court’s engagement with the conventions of constitutional amendment in the Patriation Reference and the subsequent Quebec Veto Reference illustrate the limits of its willingness, or ability, to do so even under the old, conventional regime.) And so the Supreme Court really had no choice but to try somehow to bring the less-than-fully-legal mess of Part V into the realm of legality. Inevitably, it had to do some violence to the text. It would not be fair to fault it for having done so. However, the difficulty of the Court’s position should not shield it from criticism of the way it went about its task, or absolve it from the responsibility for the problems which its endeavour will create. In particular, the concept of “constitutional architecture” which it used deserves critical attention. I hope to provide it shortly.

The Façade and the Edifice

This is my much-delayed post on the Supreme Court’s opinion, issued last Friday, in Reference re Senate Reform2014 SCC 32. Although the Court’s conclusions, all of which I had correctly predicted the day before, were not really a surprise, its reasoning was somewhat unexpected. It is also rather vague and difficult to understand. This may have been the price to pay for a unanimous decision, an attempt to paper over some underlying disagreements between the judges about the best approach to take. But the Court’s reasoning might also signal a new departure, a shift in Canadian constitutional law and theory. 

As everybody knows by now, the Supreme Court held that the introduction of term limits for Senators and of “consultative” elections to the Senate require the consent of seven provinces with more than half of Canada’s population, while the abolition of the Senate requires the unanimous consent of all the provinces. The requirement that Senators own a property worth at least 4000$ can be abolished by Parliament acting alone, except with respect to Québec, which has to consent to the modification of the special rule pursuant to which its Senators must own property within specific districts in the province.

After throat-clearing describing the Senate’s roles as a legislative chamber of “sober second thought,” and one in which both the regions of Canada and minorities, not well represented in the House of Commons, could have a voice, the substantive part of the Court’s opinion begins with a discussion of the nature of the Canadian constitution and constitutional amendment. And this is where it gets interesting.

The constitution, says the Supreme Court, includes not only texts (both those referred to in subs. 52(2) of the Constitution Act, 1982 and others), but also an “architecture,” which has to do with the way in which its components interact and complement each other. This architecture might be similar or related to the underlying constitutional principles which the Court identified in Reference re Secession of Québec, [1998] 2 SCR 217, in that like the principles, it informs constitutional interpretation. But it is a more multi-faceted concept than that of underlying principles. The “architecture” consists of “the assumptions that underlie the text and the manner in which the constitutional provisions are inteded to interact with one another” (par. 26). And, crucially, since the constitution does not only consist of text but also of its “architecture,” “amendments to the Constitution are not confined to textual changes. They include changes to the Constitution’s architecture” (par. 27).

Another, seemingly separate, consideration in understanding and applying the rules regarding constitutional amendment codified in the Constitution Act, 1982, is “the principle that constitutional change that engages provincial interests requires both the consent of Parliament and a significant degree of provincial consent” (par. 29). The

amending formula [is] designed to foster dialogue between the federal government and the provinces on matters of constitutional change, and to protect Canada’s constitutional status quo until such time as reforms are agreed upon (par. 31).

Therefore, the “7/50” amending formula (requiring the consent of two thirds of the provinces with at least half of Canada’s population), which balances flexibility and the need for consensus, is the default option. Other amending formulae are exceptions. The Court provides a few remarks on those. It observes that s. 43 of the Constitution Act, 1982, requiring federal and provincial consent, is applicable to “special arrangements” for one or several provinces, which must assent to any change to such arrangements (par. 44). The Court also notes that the unilateral amendment procedures in ss. 44 and 45 are applicable only to those changes to, respectively, the federal and the provincial constitutions which “do not engage the interests of the other level of government” (par. 48). The Court adds that

[n]either level of government acting alone can alter the fundamental nature and role of the institutions provided for in the Constitution. This said, those institutions can be maintained and even changed to some extent under ss. 44 and 45, provided that their fundamental nature and role remain intact (par. 48).

The opinion then moves on to the specific questions asked in the reference. The first set of questions concerned “consultative” elections to the Senate, the results of which the Prime Minister would be obliged to “consider”, while theoretically retaining the discretion not to recommend their winners for appointment to the Senate. The federal government argued that such elections were permissible, since their introduction did not alter the constitutional text. But that argument, says the Court, “privileges form over substance” (par. 52):

 While the provisions regarding the appointment of Senators would remain textually untouched, the Senate’s fundamental nature and role as a complementary legislative body of sober second thought would be significantly altered.

This would “amend the constitution … by fundamentally altering its architecture” (par. 54). That architecture includes the Senate’s relative independence from partisan politics, and also its role as “complementary” rather than a competitor to the elected House of Commons. Elections “would give it the democratic legitimacy to systematically block the House of Commons, contrary to its constitutional design” (par. 60). Although the Court does not explicitly mention the constitutional convention which obliges the Senate to yield to the Commons’ will, the reference is unmistakable.

Remarkably, it is only after this discussion of “architecture” that the Court turns to the constitutional text, which it says “support[s]” its conclusions, and whose “words … are guides to identifying the aspects of our system of government that form part of the protected content of the Constitution” (par. 64). The “plain meaning” (par. 67) of these words ― specifically, the term “method of selecting Senators” in par. 42(1)(b) of the Constitution Act, 1982 ― covers “consultative” elections, because they describe not only the formal mechanism whereby Senators are appointed by the Governor General, but also the “selection” of the persons thus appointed.

The second question the Court had to consider envisioned the introduction of term limits of varying length for Senators. Here, it begins with text, observing that s. 42 does not refer to term limits, but also asserting that “7/50”, not unilateral amendment, is the default procedure even for changes to the Senate not mentioned in s. 42:

Changes that engage the interests of the provinces in the Senate as an institution forming an integral part of the federal system can only be achieved under the general amending procedure (par. 75).

The Court takes the view that the imposition of term limits would fundamentally change the Senate, by limiting the Senators’ independence and thus undermining their capacity for “sober second thought.” Although a very long fixed term would conceivably be equivalent to the current system of appointment until the age of 75 in this respect, it is impossible for a court to decide what length would be sufficient; “this is at heart a matter of policy.” And thus

[t]he very process of subjectively identifying a term long enough to leave intact the Senate’s independence engages the interests of the provinces and requires their input” (par. 82).

Property qualifications, by contrast, do not engage provincial interests. Their removal would not interfere with the Senate’s role or the Senators’ independence, and can (except with respect to Québec, as explained above), be achieved unilaterally.

The final question of the reference concerned the outright abolition of the Senate. Here again, “architecture” becomes arguably the dominant factor in the analysis. Although the Court also finds that the abolition of the Senate would amend the amending formula itself, requiring unanimous provincial consent, this is not the first consideration it mentions. What comes first is the assertion that

abolition of the Senate would fundamentally alter our constitutional architecture — by removing the bicameral form of government that gives shape to the Constitution Act, 1867 (par. 97).

The Court also observes that the abolition of the Senate was not contemplated in 1982, so that it cannot be construed as a mere reduction ― to nothing ― of its “powers” and membership, as the federal government argued.

I will try to have some more specific comments on the Court’s opinion, but here are three quick ones. First, as I noted in earlier posts on this case, the Supreme Court has consistently resisted attempts at unilateral constitutional reform, whether by the federal government (starting 35 years ago in Re: Authority of Parliament in relation to the Upper House, [1980] 1 S.C.R. 54, then in the Patriation Reference, and now again), or by a province (in the Secession Reference). Whatever else it thinks of the constitution, the Court clearly believes it is one that requires consensus. Second, beyond that, it is still not really clear what the constitution is. In particular, it is not clear just what the “constitutional architecture” the Court relies on is. However, as I suggest above, it seems to include (at least some) constitutional conventions. Yet the Court simply did not address the serious (although in my view not insuperable) theoretical and practical difficulties with incorporating conventions into constitutional law. Third, and most broadly, the court seems to have given a remarkably bold answer to the question I asked in my analysis of the oral argument ― what to make of the constitution? ― or, more precisely, of the constitutional text. The text, its opinion suggests, is no more than a façade. We can look at it, of course, and it can give us useful hints of what lies behind it. But the real architecture of the constitutional edifice is hidden from our view. And if there is a door through which one can pass to observe it, it is only the Supreme Court that has the key.

It’s a Dangerous Thing…

… To make predictions, especially about the future; so Winston Churchill. But the attraction of doing so is irresistible, so here goes: my forecast for the outcome of the Senate Reference, which the Supreme Court will release tomorrow. (If you need a refresher on the Reference and the issues it raises, as well as another set of predictions, have a look at Emmett Macpharlane’s excellent “Guide to the Senate Reform Debate”; you can also read my collected posts on the topic here.)

On the question of term limits, my guess is that the Court will find that none of the limits suggested by the federal government are constitutional. Although, as prof. Macfarlane suggests, the “compromise” view defended by Ontario and Saskatchewan, according to which long term limits would not be problematic, I doubt that it can be worked into a rule that is at once clear enough and principled enough for the Court to endorse it. The federal government itself stubbornly refused to offer the Court much in the way of a principle limiting its power to impose term limits, and the provinces supporting the compromise solution have done little better. Although the constitutional text does not offer direct support for the proposition that the Senators’ term is, in fact, entrenched, the Court will hold that essential characteristics of the Senate cannot be unilaterally amended by Parliament.

I also expect the Court to reject “consultative” elections to the Senate, partly for the same reason, but also because, in one way or another (hopefully, in the way Fabien Gélinas and I suggest!), the discretionary appointment of Senators by the Prime Minister is part of the entrenched “method of selecting Senators.” The Supreme Court will not countenance the federal government’s attempt to achieve unilaterally and indirectly what the framers of the Constitution Act, 1982, quite clearly thought required a constitutional amendment with provincial participation.

However, the Court will hold that the property qualification for appointment to the Senate can be unilaterally abolished by Parliament, because such an amendment does not interest the provinces (by their own admission), and because, however important in 1867, the property qualification, eroded by inflation to a fairly nominal amount, is no longer an essential characteristic of the Upper House.

Finally, regarding the abolition of the Senate, the Supreme Court will find that unanimous provincial consent is required. Although, unlike prof. Macfarlane, I am not convinced that the contrary conclusion (that the Senate can be abolished under the general, “7/50”, amending formula), would require the Court “to divorce its reasoning completely from the constitutional text,” ― which, after all, does not list the abolition of the Senate among the matters requiring unanimity ― the judges seemed to find the prospect of such a fundamental amendment done with anything less than unanimous support quite unsettling. Although the Senate might, as prof. Macfarlane suggests, be protected from abolition by its role in the process of constitutional amendment itself, which cannot be changed without unanimous consent, the Court might even base its ruling on the broader ground that  any fundamental alterations to the nature of the Canadian constitution require the consent of all the provinces.

Well, we’ll see. I hope I don’t look like too much of an idiot tomorrow.

UPDATE: I forgot to mention that I will be at conference ― on Senate Reform ― at Université Laval’s Faculty of Law tomorrow. If you happen to be there, come say hi. In any case, I don’t think that I’ll have time for a blog post tomorrow. Except perhaps a public acknowledgement of idiocy, if necessary. I’ll do my best to have something more substantial on Saturday.

Not So Super Majorities

We all want to live under good constitutions… whatever good really means. But how do we make sure that our constitution is, in fact, good? In a post at the Volokh Conspiracy (part of a series discussing their book on originalism), John McGinnis and Michael Rappaport argue that that “stringent supe[r]majority rules provide the best way to make a national constitution.” A constitution, they say, should be enacted by a super-majority and placed beyond the reach of amendment, except again by supermajority. I do not think that they succeed in demonstrating this claim.

First, it is important to distinguish two questions which professors McGinnis and Rappaport seem to run together. One concerns the best way to enact a new constitution; the other concerns the best way to amend an existing one. To be sure, this distinction is somewhat artificial and can be difficult to make; at some point an amendment might be so fundamental as to be equivalent to the making of a new constitution. Nevertheless, there seems nothing wrong with the idea that, say, the unanimous agreement of formerly independent states is required to create a new federal constitution uniting them, which will subsequently be amendable without their unanimous consent. The opposite move ― the creation of a constitution with lower support than would be required to amend it ― might seem sneaky, but I’m not sure that it is actually wrong in all circumstances. In Canada, the Constitution Act, 1982, was entrenched with the support of nine provinces, yet it requires the agreement of all ten for some types of constitutional amendment. I am not aware of anyone having criticized it for that reason.

Be that as it may, professors McGinnis and Rappaport seem to think that both the initial creation of a good constitution and the making of good amendments thereto require supermajority agreement. They make four arguments in support of this view. First, [“s]upermajority rules … screen norms for substantial consensus and bipartisan support,” which, in turn, “creates legitimacy and allegiance as citizens come to regard the Constitution as part of their common bond.” Second, “supermajority rules … encourag[e] richer deliberation about the Constitution.” Third, supermajority requirements ensure that citizens know that they might not be able to change the constitutional rules when they apply to them, and hence are more likely to settle on rules that are fair to all and express the public interest rather than partisan advantage. And fourth, “[s]upermajority rules also generate constitutions that are more likely to protect minorities,” since minorities are able to block the enactment of rules that do not protect them.

I do not think that any of these arguments proves that supermajority rules are either necessary or sufficient to make good constitutions. To keep this post from being much too long, I will mostly focus on the first claim ― that supermajority agreement on the contents of a constitution makes it more legitimate. However, here some quick thoughts about the other ones. Supermajority rules can, indeed, encourage richer deliberation, but they can just as well encourage unprincipled compromise, such as the “trading of fish for rights” that preceded the enactment of the Constitution Act, 1982 ― or, say, the compromise that preserved slavery and even give slave-owning states disproportionate political power under the U.S. Constitution as originally enacted. For the same reason, supermajority rules do not guarantee that a constitution will respect the public interest ― they might simply encourage horse-trading between partisan factions instead. And while supermajority rules can protect minorities, they can also give them the power to extract disproportionate advantages or protections that are larger than necessary and unduly impede collective action.

To return now to the claim that the legitimacy of and citizens’ loyalty to a constitution depend on the breadth of the consensus on its contents, it seems to me that it leads to absurd consequences. Most obviously, it means no only that supermajority is better than simple majority, but also that unanimity is better than any other supermajority. Yet professors McGinnis and Rappaport are not arguing for unanimity. Perhaps that is merely because it would be impractical. Perhaps also there is, in fact, a diminishing return on additional support, at least past a certain threshold. Yet it is not clear where that threshold lies, and whether it does in fact lie in supermajority territory. For some purposes―including elections in the Canada as well as in the U.S. ―we accept a plurality, not even a simple majority as sufficient for the win. Indeed, it is possible for a party or a candidate to win such elections without even a plurality of the national popular vote. It is reasonable to demand that a constitution, expected to endure for decades and even centuries, enjoy higher support than a politician elected to hold office for four or five years, but it is by no means clear just how much higher.

However, there is an even more fundamental problem with the argument that constitutions are best enacted and amended by supermajorities because their legitimacy is a function of the consensus they generate. If a constitution (however enacted) can only be amended by a supermajority, then it is quite possible for a constitutional provision to remain in force even though a majority of the people come to think that it is a rotten one. The more stringent the supermajority requirement, the wider consensus the can be on the need for constitutional amendment without the amendment being realized. From the standpoint of a consensus on the substance of constitutional provisions, the constitution is in such circumstances quite illegitimate, and thus supermajority requirements make illegitimate constitutions more, rather than less, likely.

I am inclined to think that what makes a constitution ― or, for that matter, a government ― legitimate is not the breadth of substantive agreement with it, but agreement on the procedures that led to its creation. This is what explains the continuing legitimacy of constitutional provisions with which a majority of citizens ― but one not sufficient to overcome a supermajority requirement ― disagree with. The focus on the contents of constitutional rules imposes too high a threshold for their legitimacy.

What I have said so far means that supermajority rules are not sufficient to make a good constitution. They are also not necessary. It is, surely, possible to arrive at good rules (whether by this we mean rules that generate broad agreement, or are conducive to the public interest, or protect minorities) by other procedures too, whether simple majority, through the development of constitutional conventions, or through adjudication. (On this last point, professors McGinnis and Rappaport recognize, in a subsequent post, that in some cases, judicial “precedent now enjoys such strong support that it is comparable to that necessary to pass a constitutional amendment.”)

The question of what procedure, if any, is the best for making a good constitution is clearly a difficult one. I have not attempted to answer it in this post. I do think, however, that professors McGinnis and Rappaport have not succeeded in demonstrating that supermajority enactment and amendment is that procedure. Supermajority requirements have some benefits, but also considerable flaws.

What to Make of the Constitution

I have written a post on the Senate Reference hearings for I-CONnect, the blog of the International Journal of Constitutional Law. In large part, it follows up on and develops some of the ideas I had in my first impressions post last week, with a bit more context. I am cross-posting it below.


Over the course of three days last week, the Supreme Court of Canada heard submissions from the federal government, the ten provinces, two territories, two ami curiae, and several interveners on the constitutionality of the federal government’s proposals for reforming the unelected upper house of the Parliament of Canada, the Senate. Beyond the obvious importance of possible Senate reform, which has been discussed almost as long as Canada has existed, this case is potentially of great significance because it is the first time the Court considers the meaning of Part V of the Constitution Act, 1982, which includes Canada’s plethora of procedures for constitutional amendment.

The questions the Court is called upon to answer concern four possible reforms. First, the limitation of the terms of a senator’s office, currently held until a retirement age of 75, to a fixed number of years (possibly 9 or 10) or to the life of two Parliaments (about 8 years under normal circumstances, but shorter in the case of minority governments). Second, the creation of a system of consultative elections to be held in the provinces that desire it, the winners of which the Prime Minister would be required to “consider” recommending to the Governor General for appointment to the Senate. Third, the removal of the “property qualification” of $4000 to hold senatorial office, a considerable sum in 1867, meant to ensure the Senate’s position as the House of privilege, but by now a somewhat comical relic of an age long gone and little lamented. And fourth, the outright abolition of the Senate.

The first three of these reforms, the federal government argued, could be implemented by Parliament acting unilaterally, either under s. 44 of the Constitution Act, 1982 (for term limits and the property qualification) or under the residual “peace, order and good government” power of s. 91 of the Constitution Act, 1867 (for consultative elections). As for abolition of the Senate, the federal government took the position that it could be done under the “general” amending formula of the Constitution Act, 1982, requiring the consent of 7 provinces representing at least 50% of Canada’s population (the “7/50 formula”). Most provinces, as well as the interveners, however, are of the view that the instauration both of term limits and of consultative elections can only be done under the 7/50 formula, and that abolishing the Senate requires the provinces’ unanimous consent.

Legally, the reasons for these disagreements come down to differences over the proper way of interpreting Part V of the Constitution Act, 1982. Part V applies to changes to “the Constitution of Canada,” and provides several different amending formulae, ranging from unilateral action to agreement between the federal Parliament and all the provinces, depending on the subject of the proposed amendment. The trouble is that the subjects singled out for special amending formulae are described in confusing terms, making it unclear under which provision of Part V a proposed amendment falls. But the confusion goes even deeper, because the very definition of “the Constitution of Canada,” to the amendments to which Part V applies, is unclear and incomplete.

Over the course of arguments before the Supreme Court, there emerged three interpretive approaches that the Court might adopt. The federal government’s preferred approach is―its denials notwithstanding―literalist. On this reading, the “powers of the Senate” entrenched by s. 44 do not include any senatorial independence that might be compromised by shortened term limits. The expression “method of selecting senators,” in the same provision, refers only to the formal appointment by the Governor General, and not to the fact that Senators are understood to be appointees lacking electoral legitimacy. And, since the list of constitutional changes requiring unanimous agreement of the provinces in s. 41 of the Constitution Act, 1982, does not specifically mention the abolition of the Senate, it does not cover it.

A second approach, championed by Québec, might be called originalist, in that it focuses on preserving the bargain struck at confederation in 1867. The Senate was, all agree, a central element of that bargain, so it is not to be touched without the consent of the parties to that bargain―that is to say, the provinces as well as the federal government. And even if some of the expectations of these parties, such as those regarding senatorial independence, have not been fulfilled, they must be upheld “regardless of reality.”

Most provinces, however, as well as the interveners and the amici curiae, urged the Court to take what was often described as the “living tree” approach, focusing not so much on the words of the constitutional text or the historical compromises that gave birth to it as on what matters most for its present and future operation. So the “method of selecting senators” is not only the part of the selection process described in the constitutional text (i.e. the appointment by the Governor General); “the powers of the Senate” include its independence, whatever exactly that might be (since it is clear that the Senate is, in fact, a political, partisan body); and the requirement of unanimous consent applies not only to the changes listed in s. 41, but also to other fundamental changes to the constitution.

In reality, however, interpretive approaches are not so neatly distinct. In what an intervener described as a dangerous application of the living tree approach, the federal government urged the Court to allow Parliament to bring the Senate into the 21st century, and not to require unrealistic levels of political agreement, condemning the country to another “135 years of talks.” Conversely, those opposed to the federal government’s proposal made much of the fact that the abolition of the Senate was simply not contemplated when the amending formula was being drafted in 1981, arguing that this, rather than the relative unimportance of such a change, explained its conspicuous absence from s. 41. As a bemused Justice Rothstein put it to Manitoba’s lawyer, “whatever works?” “Welcome to the law,” she replied.

What works, and what doesn’t, are questions that will weigh heavily on the Court’s collective mind. What works with an amending formula which, although drafted merely a generation ago, seems to reflect a very different country than the one in which we now live? A country where constitutional negotiations were not being described as the opening of a “can of worms,” and where agreement on substantive constitutional change seemed within reach; a country much preoccupied by the then-Prime Minister’s suspected republicanism, but not all by the prospect of the abolition of the Senate; a country that thought nothing of ruling its northern territories like colonies, and its First Nations like subject peoples, whose views on constitutional change could safely be ignored. What sort of authority can a constitutional text drafted under such circumstances have? Yet saying that this text is insufficient to meet the needs of today―never mind those of the years and decades to come―only forces one to confront the question of what sort of authority the Supreme Court has effectively to re-write it.

Yet the Supreme Court of Canada has not been shy about intervening in constitutional reform in the past. It stopped possible Senate reform in its tracks with the in 1979, prevented unilateral Patriation of the constitution in 1981, and ruled that a unilateral secession of Québec would be illegal in 1998. In none of these cases was it relying on explicit constitutional provisions. In all of them, it ruled that constitutional change required a broad agreement between the federal government and the provinces, and could not result from unilateral action. Although it is, of course, dangerous to make such predictions on the basis of questions asked at oral argument, insofar as it concludes that Part V of the Constitution Act, 1982 does not clearly reflect this preference for consensus, the Court did not seem very likely to heed the federal government’s call for it not to treat the constitutional text as a mere “suggestion box.”

First Impressions

I will have more structured and fuller thoughts in the next days, but here are some initial impressions, in no particular order, of the Supreme Court’s hearings on the Senate Reference, almost all of which I watched.


This case is, of course, largely about constitutional interpretation. (Much more so, in fact, than most constitutional cases in Canada, certainly much more than the many Charter cases where the constitutional text is barely alluded to.) Now, in Canada, we often think of constitutional interpretation as involving a choice between two options: the “living tree” approach on the one hand, and something like originalism on the other. But the arguments in the Senate reference suggest that there are, in fact, at least three different approaches: the living tree, which most provinces invoked; originalism, in the sense of giving an overriding importance to the bargains that made confederation possible, which was Québec’s preferred approach; and literalism or textualism, which is the federal government’s. In reality, things are even more complicated, because the provinces also rely on constitutional structure, and the amici curiae focus primarily on it. And provinces, despite their invocations of the living tree, don’t hesitate also to appeal to the original bargain. In the funniest exchange of the three days, Justice Rothstein summed it all up as “whatever works,” to which Heather Leonoff, Manitoba’s lawyer, replied: “Welcome to the law!”


Speaking of the lawyers, it’s a tough job they had. None more so than Nancy Brown, for British Columbia, the first to argue yesterday. She was, perhaps, a bit slow to get into the substance of her argument ― as all of her colleagues who preceded her, none of whom could resist the temptation of reminding the court of the historic significance of the case, had been. As Ms Brown was speaking of the importance of confederation, the Chief Justice snapped that the Court had already heard about it many times, and demanded to be told something “new.” The Chief Justice achieved what I suppose was her aim ― the lawyers who went after Ms Brown focused and shortened their arguments, trying not to repeat too much of what others had said before. But I had the impression that Ms. Brown was somewhat thrown off course by the Chief Justice’s comments, and that it was not very fair to take the court’s (admittedly understandable) frustration out on her in this way.


Not everybody had such a hard time, though. The academics involved ― Sébastien Grammond, the Dean of the Civil Law Section at the Faculty of Law of the University of Ottawa, was impressive in his argument for the Fédération des communautés francophones et acadienne du Canada, and McGill Law’s Dean Daniel Jutras, even more so. (I’m biased, of course, as an alumnus!) Dean Jutras’ fellow amicus, John Hunter was very good as well, though the Court gave him a hard time. But perhaps the most impressive performance was that of Senator Serge Joyal, who intervened and represented himself. It is often said that a lawyer who represents himself has a fool for a client, but Mr. Joyal is no fool. He spoke with both emotion and intelligence, and brought insights to the Court which the judges had been looking for (about the pre-Patriation constitutional negotiations, in which he was very much involved). Indeed, Mr. Joyal, I think, is precisely the sort of person of whom the Senate was meant to consist. Too bad there are not more of them there.


As for the best moment involving a judge, it came this morning when Justice Cromwell asked Mr. Hunter, who was arguing that the federal government’s plan to institute purportedly non-binding elections to choose persons whose appointment to the Senate the Prime Minister would have to “consider” did not entail a modification of the constitution, whether the government could, instead of elections, auction off Senate seats ― provided the auction was non-binding. (Mr. Hunter initially resisted, but eventually conceded this would be possible, subject of course to an eventual judgment of the voters. Not a great moment, I think, for the position he was supporting.)


This brings me to way this will go. The federal government is in trouble. It is, of course, dangerous to make predictions based on hearings, but I have a strong impression that the Court ― all eight judges ― was skeptical of its position. The questioning both of its lawyers and of those who supported its position (as Alberta and Saskatchewan largely did, and Mr. Hunter did in part) was tougher than that of its opponents. Especially issue of the abolition of the Senate, I would be very surprised indeed if the Court were to endorse the federal government’s view that it could be effected without the unanimous consent of the provinces.


If that is right, however, the Court may well ― though perhaps it will not ― more or less re-write the Constitution Act, 1982. The constitutional text does not, after all, mention the abolition of the Senate, which would at first glance suggest that the federal government’s position that the general amending formula, requiring the consent of 7 provinces with 50% of Canada’s population, should apply to it. What became quite clear in the course of the argument, however, is that the constitutional text does not contemplate a great many things which were not thought of in 1981, including some very fundamental changes, and that the Court seems reticent to hold that such changes (including, for example, turning Canada into a dictatorship), might happen with less than unanimous support. Judging by the tenor of the arguments, the Court may well follow the suggestion of some provinces and of the amici curiae, and hold that such fundamental changes require  unanimity, regardless of what the constitutional text says ― or, rather, doesn’t say. The federal government’s lawyers’ concluding exhortation not treat the Constitution Act, 1982, as a “suggestion box” may have come too late. Yet there would have been no need for it if the government itself had not done exactly that with its plan for getting to an elected Senate through the back door.


Whichever way the Court’s decision goes ― whether it ends up endorsing the federal government’s plan for getting around constitutional text or re-writing that text ― it would be pretty revolutionary. The Chief Justice might have been tired of hearing it, but this is, indeed, a historic case.

What We Said

Apologies for the recent silence. There was no particularly good reason for it, either. Anyway, I’m back.

And there is a very good reason for that: the Québec Court of Appeal has released its opinion in response to a reference by the Québec government on the constitutionality of the Federal Government’s Senate reform plans, which involve the limitation of Senators’ terms to 9 years and, more importantly, the setting up of provincial elections the appointment of the winners of which a Prime Minister would be obliged to “consider” recommending to the Governor General. In Reference re Bill C-7 Concerning the Reform of the Senate, 2013 QCCA 1807 (the French opinion is here; English translation here), the Québec Court of Appeal holds that this project is unconstitutional if undertaken unilaterally by Parliament; it can only be implemented as a constitutional appointment pursuant to par. 42(1)(b) of the Constitution Act, 1982.

The Court starts off by explaining the importance of the Senate to the Fathers of Confederation. The constitution of Canada was meant to be, as the Preamble put it, “similar in principle to that of the United Kingdom,” and that meant, among other things, having an unelected Upper House of the legislature. The province of Canada had, in fact, experimented with an elected Upper House ― and Sir John A. Macdonald had not liked the experience. The appointed, undemocratic Senate was an essential part of the bargain struck in 1867. To this day, “it seems that the Senate and its members play a significant role in federal political life, and that the institution is not simply a mirror of the House of Commons” (par. 12). The Supreme Court has confirmed the Senate’s position as an entrenched, central part of the compromise that made Confederation possible, in Re: Authority of Parliament in relation to the Upper House, [1980] 1 SCR 54.

So much for the context. The relevant constitutional text consists of, on the one hand, par. 42(1)(b) and, on the other, s. 44 of the Constitution Act, 1982. The former provides that an amendment according to what is usually referred to as the 7/50 procedure, requiring the consent of 7 provinces representing between them at least half of the Canada’s population, is necessary to effect “[a]n amendment to the Constitution of Canada in relation to” any of a number of “matters,” among which are “the powers of the Senate and the method of selecting Senators.” As for s. 44, it provides that “[s]ubject to sections 41 and 42, Parliament may exclusively make laws amending the Constitution of Canada in relation to the executive government of Canada or the Senate and House of Commons.” Also relevant are s. 24 of the Constitution Act, 1867, which provides that Senators are “summon[ed]” by the Governor General, and par. 41(a) of the Constitution Act, 1982, which requires the unanimous consent of the provinces to amendments “in relation to … the office of the … Governor General.” Québec argued that the proposed Senate reform came within the terms of par. 42(1)(b) as affecting “the method of selecting Senators” and/or par. 41(a) as affecting the office of the Governor General. The Court accepted the former claim, and rejected the latter.

S. 42, it said, should not be interpreted restrictively, as an exception to a more general principle contained in s. 44. These provisions are of equal importance. Amendments relating with the “internal management” of the Senate fall under s. 44; those that have to do with the Senate’s “role[s] within the federal legislative structure, in particular those of ensuring provincial and regional representation and examining bills with sober second thought,” under s. 42 (par. 34).

Crucially, Parliament cannot get around the entrenchment of s. 42 by legislating so as to leave in place the formal provisions of the Constitution while changing the way it operates in practice. For one thing, this would contradict “the principle of supremacy of the Constitution” (par. 43). For another, it would subvert the compromise that made possible the Patriation of the 1981/82, which, so far as the Senate was concerned, had consisted in kicking the can down the road, and postponing any amendments ― to be effected at some later date pursuant to the new amending formula. Finally, s. 42 must be interpreted in light not only of the legal formalities, but also of the political realities of the constitution:

section 42 cannot be read as reflecting a consensus between the federal and provincial governments in 1982 to preserve the formalism but not the reality with respect to the matters set out therein, including the method of selecting senators. … [W]hat interest would the provinces have had when the Constitution Act, 1982 was adopted to protect a juridical reality that, even then, was inconsistent with political reality?

The political reality is that “the method of selecting Senators,” as it existed in 1982, included no electoral process. “The method of selecting Senators” refers not only to their final appointment by the Governor General, but to the entire process leading to that appointment. That process would be modified by the federal government’s reform project. Therefore that project requires a constitutional amendment.

That amendment need not be unanimously supported by the provinces, however, because it does not affect “the office of the Governor General”. While the Governor General is responsible for the final appointment of the Senators as a matter of law, “[i]n reality, the appointment of senators became the exclusive prerogative of the Prime Minister who was then in office whenever a vacancy occurred” (par. 55). The federal government’s reform project would have affected not the Governor General’s (purely formal) role in the process, but the Prime Minister’s. And “to assimilate an amendment of the powers of the Prime Minister with those of the Governor General for the purposes of paragraph 41(a) of the Constitution Act, 1982 would limit Parliament’s powers because of a constitutional convention. Such a limitation does not exist, or at a minimum, does not concern the courts” (par. 58). Conventions exist in a separate, non-justiciable realm. They can be modified by the behaviour of political actors; therefore, a fortiori, they can be modified by statute, without the need for a constitutional amendment.

If this all sounds familiar, it might be because you’ve read Fabien Gélinas’s and my paper on “Constitutional Conventions and Senate Reform,” in which we argued that the amending formula of the Constitution Act, 1982, must be understood in light of the constitutional conventions which determine the practical operation of the constitution. This means, on the one hand, that the “method of selecting Senators” means discretionary decision-making by the Prime Minister and no electoral process, and on the other, that “the office of the Governor General” does not in fact include the power to choose Senators. As a result, the federal government’s reform project comes within the scope of par. 42(1)(b), but not 41(a).

I am very happy with this opinion. I hope that the Supreme Court, which is set to hear the arguments on the federal government’s own Senate reform reference in a few weeks, comes to similar conclusions (and perhaps even spares a few words for us)!