Printer Friendly

EXCLUSIVELY YOURS: RECONSIDERING INTERJURISDICTIONAL IMMUNITY.

I. INTRODUCTION

A little over ten years ago, in Canadian Western Bank, (1) the Supreme Court of Canada, alarmed by what it called "the dangers of allowing the doctrine of interjurisdictional immunity to exceed its proper (and very restricted) limit", (2) set about reducing access to it substantially.

Before then, it was understood "that each head of federal legislative power under the Constitution Act, 1867, possesses basic, minimum, and unassailable content, which the provinces are not permitted to regulate indirectly through valid laws of general application" and that " [t]his exclusive federal jurisdiction precludes the application... of provincial statutes which have the effect of regulating" anything comprising that "core" of federal legislative authority. (3) But according to the majority of the Court in Canadian Western Bank, interjurisdictional immunity (IJI), though "in principle applicable to all federal and provincial heads of legislative authority,"' "is of limited application and should in general be reserved for situations already covered by precedent." (5) "[T]he absence of prior case law favouring its application to the subject matter at hand", the majority added, "will generally justify a court proceeding directly to the consideration of federal paramountcy", (6) postponing the IJI inquiry, if not omitting it altogether. Finally, under this new dispensation:
[I]t is not enough for the provincial legislature simply to "affect"
that which makes a federal subject or object of rights specifically of
federal jurisdiction. The difference between "affects" and "impairs" is
that the former does not imply any adverse consequence whereas the
latter does.... It is when the adverse impact of a law adopted by one
level of government increases in severity from "affecting" to
"impairing" (without necessarily "sterilizing" or "paralyzing") that
the "core" competence of the other level of government (or the vital or
essential part of an undertaking it duly constitutes) is placed in
jeopardy [under IJI], and not before. (7)


Why the facts of that particular case invited this response is still a mystery. The Court could have reached its preferred result there--concluding that banks, when selling insurance, are subject, like everyone else in the province, to Alberta's insurance laws--without departing at all from previous understandings of IJI and its scope. Bastarache J, in concurring reasons, did exactly that. (8) But for whatever reason, the Court chose that occasion to curtail future use of IJI. (9) Subsequent decisions have embraced the new paradigm, (10) reiterating the Court's reservations about the doctrine (11) and reminding the reader, even on the two occasions since Canadian Western Bank on which the Court has invoked it, (12) that IJI can be dangerous unless used precisely as directed. For mature audiences only; parental guidance strongly recommended.

What is all the fuss about? And what happened in 2007 to make IJI seem so much more threatening than it had generally seemed before? (13) I don't know the answer to the second question; it's interesting, though, that despite having disapproved and narrowed previous IJI doctrine, the Court continues to say that IJI remains available in situations already covered by precedents drawn from that earlier doctrine. (14) As to the first, in Canadian Western Bank and its progeny, the Court has offered six reasons why we need to handle IJI with particular care. IJI is said to: (1) be superfluous, because the paramountcy doctrine gives Parliament all the power it needs to neutralize intrusive provincial legislation; (15) (2) risk creating "legal vacuums"; (16) (3) be in tension with the principle of cooperative federalism, whereby both orders of government work rogether to address problems of common interest; (17) (4) risk giving effect to an "unintentional centralizing tendency in constitutional interpretation"; (18) (5) be out of step with the "dominant tide" in modern division of powers jurisprudence, which prefers to tolerate, where possible, the concurrent operation of legislation from both orders of government; (19) and (6) assume the existence of abstract "cores" of exclusive legislative authority whose postulation "is not compatible, generally speaking, with the tradition of Canadian constitutional interpretation". (20) In brief, this jurisprudence suggests, IJI is inconvenient, incorrect (politically and methodologically), and unnecessary. If we're not quite ready to discard IJI altogether, we ought (the argument runs) to keep it safely and securely bottled, reserving ir for exceptional use and even then, only where the situation is serious.

Personally, I don't find any of this especially convincing. In my judgment, the reasons the Court has offered for restricting resort to IJI are unpersuasive and the restrictions imposed, therefore, artificial. I propose in this article to say why, but more importantly, to revisit the very idea of IJI as a feature of the Canadian constitutional order. If we are to make fully informed decisions about IJI, we need to understand better than it seems we currently do how it works and what it contributes to Canadian federalism.

It is true that IJI can seem esoteric, even counterintuitive. It is fair for one to wonder what useful purpose it still serves. To answer that question, we need to begin by reacquainting ourselves with the reasons we have been taught to care about Canadian federalism and to satisfy ourselves whether they still matter. I argue in Part Two that they do. A functioning federal system requires a means of ensuring that each of its component orders of government minds its own business. Then we need to turn our attention to IJI itself: to what distinguishes it from other, related constitutional doctrines; to whether (as some have suggested) it's reciprocal, protecting exclusive provincial as well as exclusive federal authority (short answer: probably not); and to IJI's particular contribution to our constitutional scheme. These are the subjects of Part Three. My view is that IJI deserves a place within that scheme, because it is all that protects a crucial but under-recognized aspect of parliamentary sovereignty: Parliament's freedom to legislate, or not to legislate, in respect of matters within its exclusive legislative authority. Were it not for IJI, only the federal government would have any capacity to limit the impact on exclusively federal matters of otherwise valid provincial legislation.

Part Four considers and rejects each of the six reasons the Supreme Court has offered for circumscribing dramatically the reach of IJI doctrine. Part Five engages the underlying normative inquiry and suggests an approach to IJI more resonant than the current one with the constitutional values that support the doctrine's retention. I argue, in brief, that IJI should protect all but only those matters within federal legislative authority deemed to have sufficient national importance to need protection from collateral as well as deliberate provincial supervision. A matter will have sufficient national importance to attract

IJI when the national interest, or fidelity to the purpose underlying the grant of specific federal authority, requires a measure of consistency or coordination of approach in respect of it nationwide. This will be so when it would compromise the national interest, or realization of that purpose, to leave the matter vulnerable to legislation reflecting the interplay of (otherwise perfectly legitimate) local or regional preferences or concerns. The kinds of provincial measures against which IJI should immunize are those whose unintended effects on such matters would, if intended, have sufficed to render those measures invalid. Part VI contains some concluding remarks about IJI.

II. FEDERALISM 101

In the Quebec Secession Reference, (21) the Supreme Court put federalism first on its list of the "four foundational constitutional principles" that comprise "the vital unstated assumptions upon which the text [of Canada's Constitution] is based." (22) "It is undisputed", the Court insisted, "that Canada is a federal state"; (23) "there can", it added, "be little doubt that the principle of federalism remains a central organizing theme of our Constitution." (24) The Court in Canadian Western Bank agreed. (25)

But federalism, pared to its essence, requires only that there be more than one independent source of authority within a given polity. In Canada, "political power is shared by two orders of government: the federal government, on the one hand, and the provinces, on the other." (26) Strictly speaking, that is all one need say to affirm that Canada is a federal state. Federalism does not require any particular allocation of the powers available to its constituent orders of government. Different federations allocate powers differently.

So here is a thought experiment. What would be wrong with saying that our two orders of government, federal and provincial, could each make laws about whatever they wanted, subject only to the constraints of the Charter, (27) section 35 of the Constitution Act, 1982, (28) and sections 96 (29) and 121 (30) of the Constitution Act, 1867? As long as the Constitution protects these fundamental rights from unjustified government interference, protects a guaranteed core of superior court jurisdiction altogether, and prohibits intentional restrictions on interprovincial trade, what difference does it make which order of government legislates what?

Well, to start with, it might matter to the constituent orders of government themselves. If the federal order in a federated arrangement were going to have any meaningful authority to govern nationwide, at least some federal enactments would need priority (paramountcy) over provincial enactments, in case of conflict. Otherwise, there would be serious risk of confusion, fragmentation, or both. But federal paramountcy, coupled with otherwise unconstrained federal authority, puts all provincial arrangements and interests continually at risk. (31) From a typical provincial standpoint, therefore, a federal system featuring a federal paramountcy rule requires that there be clear, enforceable limits on the scope of the federal authority. (32) By the same token, there must be at least some constraints on the scope of provincial authority if the federal state is to have a meaningful national presence and identity. If each province had its own embassies in foreign capitals, its own currency, its own unique criminal law, and its own militia, for example, it might be difficult to conclude that they were in any meaningful sense united. To describe these several provinces as components of a single federal state is to presuppose shared acceptance that at least some coordination and consistency at the national level is necessary and appropriate.

A functioning federal system, therefore, appears to need at least some constraints on the scope of both provincial and federal authority, (33) and some mechanism independent of both for enforcing these constrainrs when one order of government perceives the other to be invading its space. (34) But why, if at all, does the conceptual geography of federalism matter to the rest of us? As long as the Constitution protects our Charter rights, the particular rights of Indigenous peoples, and the special role of the superior courts and precludes deliberate barriers to interprovincial trade, what business is it of everyday Canadians which order of government does what? If neither order of government objects to what the other is doing, why is there even a problem?

Frequently in recent years, the Supreme Court has noted with interest federal support for provincial enactments under challenge before it. In Schneider, the majority described such support as a "factor which plays no part in the determination of the constitutional validity of the Act, but which, as a practical matter, is not negligible". (35) Five years later, in OPSEU, Dickson CJC, concurring in the majority result, acknowledged that "the fact of federal-provincial agreement on a particular boundary between their jurisdictions is not conclusive of the demarcation of that boundary", but added that "the Court should be particularly cautious about invalidating a provincial law when the federal government does not contest its validity or, as in this case, actually intervenes to support it". (36) Most recently, in Kitkatla, a case about Indigenous interests in which "the Attorney General of Canada has intervened in support of the view of the British Columbia government with respect to the latter's right to legislate in this area", a unanimous court, adopting Dickson CJC's remarks on this issue in OPSEU, said that such support "does invite the Court to exercise caution before it finds that the impugned provisions of the Act are ultra vires the province." (37)

Federal-provincial concurrence about the permissibility of particular legislation is, in other words, presumptively persuasive in its favour. But as these quotations indicate, the Court has stopped short of deferring altogether to such concurrence. In 2006, it found a provincial statutory provision constitutionally inapplicable despite combined provincial and federal support for its application. (38) But if both orders of government agree that a given exercise of legislative authority is constitutionally legitimate, why should anyone else have reason to object? Why should federal-provincial concurrence on a division of powers issue not be dispositive?

The usual answer is the one that Dickson CJC himself supplied in OPSEU, in the very passage quoted with approval in Kitkatla: "The distribution of powers provisions contained in the Constitution Act, 1867 do not have as their exclusive addressees the federal and provincial governments. They set boundaries that are of interest to, and can be relied upon by, all Canadians." (39) To similar effect in a much earlier judgment:
The Constitution of Canada does not belong either to Parliament, or to
the Legislatures; it belongs to the country and it is there that the
citizens of the country will find the protection of the rights to which
they are entitled. It is part of that protection that Parliament can
legislate only on the subject matters referred to it by section 91 [of
the Constitution Act, 1867] and that each Province can legislate
exclusively on the subject matters referred to it by section 92. (40)


Put differently, we--you and I, individually and collectively--have, under this reasoning, an implied constitutional right to judicial protection from legislative measures enacted by the wrong order of government. But it makes conceptual sense to speak of such a right only if we're prepared to accept that there is such a thing as "the wrong order of government" ascertainable independently of the politics of the moment: that it really does, in other words, make a meaningful difference, overall and day to day, which order of government does what. And this will be so only if we have good reason to respect not only the fact that our Constitution distributes authority but the way in which it does so. Random or arbitrary constraints on legislative authority would afford no effective protection, serve no useful purpose. They would deserve enforcement, if at all, only as between the contestant orders of government themselves.

If everyday Canadians--those of us who are not ourselves orders of government--have reason to care about the division of powers, then, and to be able to enforce it, it is because we share the view that our Constitution has distributed powers in a way that is worthy of respect and protection. There are two clear reasons why we might reasonably hold this view.

First, this particular distribution of legislative authority is of the essence of the Confederation undertaking (the business we have chosen, so to speak). It was on the basis of just this distribution of authority that those in the original federating provinces agreed to unite into a single dominion; (41) it was to just this distribution of authority, in relevant respects, (42) that those representing the provinces subsequently joining Confederation agreed. Other things being equal (and pending approved amendment of the Confederation arrangement), it seems reasonable that we be able to count on ongoing compliance with that distribution.

Other things might not be equal, of course, if it were clear that the division of powers comprising Canadian federalism were counterproductive or irrational; reasonable people can, and do, differ on whether it is optimal, but that is a separate matter. (43) But the Court confirmed again recently that "[s]ections 91 and 92 of the Constitution Act, 1867 assign each power to the level of government best suited to exercise it". (44) And there is, at an absolute minimum, a kind of rough rationale to Canadian federalism. According to the Court in Moloney, "[b]road powers were given to the provincial legislatures with respect to local matters, in recognition of regional diversity, while powers relating to matters of national importance were given to Parliament, to ensure unity". (45) Albert Abel (46) and John Saywell (47) have characterized the catalogues of federal and provincial powers in greater detail along essentially similar lines. "If the bits and pieces of an intricate construction like the [Constitution Act, 1867] are not to be displaced or distorted to their utter confusion," Abel has argued, "the whole project must be assumed to have been, in the minds of its authors, harmonious and consistent". (48) According to Saywell, such an assumption has historical anchorage:
There had been surprisingly little discussion of the enumerated heads
or classes of subjects at the Quebec Conference or during the debate in
the Canadian Assembly. As the jurisdictional allocation had been
crafted by the Canadians, it obviously reflected their history and
mirrored the broad terms on which the English and French majorities
could live together in one state and apart as distinct societies. (49)


If this, or something like it, makes sense to you, both as an accurate representation of the Canadian federal framework and as a potentially coherent and workable way of differentiating the proper business of our two codified orders of government, then you have an interest in ensuring--pending, again, deliberate constitutional amendment--that each order stick to its own proper business and that you have a way, in principle, of invoking discipline againsr them when they do not. The Supreme Court itself has recognized this interest and taken it seriously, denying effect to attempts by either order of government to impose precondirions on proceedings that seek to invoke such discipline. (50) We now must examine carefully the mechanisms appropriate to such discipline.

III. UNDERSTANDING IJI

Acknowledging that the division of powers--the Confederation arrangement--still warrants presumptive judicial enforcement, even, where appropriate, at the behest of non-governmental litigants, leaves unanswered how and when such enforcement should occur. These questions bring IJI into the conversation.

In respect of division of powers enforcement, two propositions seem still to command virtually unanimous assent. The first is that neither order of government may set out to do something in legislation that is the proper business of the other order of government. If a federal statutory measure proves, under scrutiny, to be really about (to have, in the unforgiving parlance, as its pith and substance) something that is the proper subject only of provincial legislation, or vice versa, that measure is constitutionally invalid, wholly without legal force or effect. In determining what a given legislative measure is really about, the courts look principally at its purpose; in ascertaining its true purpose, however, they also take account of its effects. Suspiciously disproportionate (colourable) effects on persons, places, or things that are subject exclusively to the legislative authority of the other order of government can suggest an inappropriate surreptitious purpose and disqualify legislation that might in form look innocent. (51) Put differently, the Constitution does not permit either order of government to achieve indirectly what it lacks the authority to achieve directly. (52)

The other uncontroversial feature of Canadian federalism is federal paramountcy: the requirement that valid provincial legislation operate compatibly with valid federal legislation, neither conflicting in operation with nor frustrating the purpose underlying the federal law. Where there is incompatibility, the federal legislation prevails and displaces the provincial law (in the parlance, renders it inoperative), but only to the extent of the incompatibility and only while the incompatibility subsists. (53)

I have, of course, oversimplified things somewhat. Despite agreement about the ongoing importance of the inquiries into validity and paramountcy, there has been, and continues to be, dispute about the most appropriate ways of conducting them. (54) These complications need not concern us here. The key question for our purposes is whether these two inquiries, taken together and undertaken appropriately, suffice to police and preserve the distribution of legislative authority that will comprise, until further notice, Canadian federalism. If the answer is yes, there is no need, if there ever was, for IJ. We'll be better positioned to answer that question after reviewing what IJI is and what contribution it makes to Canadian federalism.

A. WHAT IS IJI?

IJI, the Court has said, is a doctrine that pertains particularly (though, as it turns out, not only) (55) to certain persons (e.g. Indians), places (e.g. lands reserved for the Indians) or things (e.g. federal undertakings or federally incorporated companies) among the enumerated classes of subjects that the Constitution assigns exclusively to federal legislative authority. (56) At a minimum, IJI protects absolutely matters relating to these subjects from unacceptable provincial (57) interference--the current standard, again, is "impairment" (58)--"in what makes them specifically of federal jurisdiction." (59) "This is so," the Court has said, "because these specifically federal aspects are an integral part of federal jurisdiction over such things or persons and this jurisdiction is meant to be exclusive." (60)

IJI--an inquiry into the applicability of certain provincial legislation--differs importantly from both the inquiry into the validity of legislation and the inquiry into its compatibility with valid federal law (paramountcy). To begin with, it does not impugn the validity of the relevant provincial legislation. Instead, it restricts the constitutionally permissible reach of provincial legislation that the courts have already determined to be valid: legislation that is, as it should be, really about something within the proper business of the provincial order. It reads down such legislation to preserve its validity, having accepted that the province intended not to encroach on exclusive federal authority. In other words, IJI protects certain core federal matters--matters deemed to come exclusively within federal legislative authority--even from inadvertent impairment resulting from legitimate provincial legislative activity. It precludes the provinces from doing inadvertently, carelessly, or accidentally what the Constitution prohibits them from doing advertently. (61) Secondly, IJI operates even in the absence of any conflicting federal law--even where the federal order has not chosen, or has chosen not, to exercise its own legislative authority in relation to those core federal matters. (62) This differentiates IJI from the paramountcy doctrine.

B. DOES IJI WORK BOTH WAYS?

On several recent occasions, the Supreme Court has suggested that IJI is reciprocal: that it can also protect from federal impairment matters within exclusive provincial legislative authority. (63) Examples of such reciprocity are scarce. Canadian Western Bank offers three, (64) only one of which, Dominion Stores, seems at all apt for the purpose. (65) Robin Elliot provides a couple of others. (66) Clark v Canadian National Railway Co (67) and hen v Simms, (68) he argues, "add force to the thesis that the doctrine operates both ways." (69)

At issue in Dominion Stores was an offence provision in federal legislation prescribing voluntary minimum grading standards for certain agricultural products and prohibiting anyone from applying these grade designations to products that failed to meet the minimum standards. The Court did not determine the overall validity of the federal offence provision but said that it "has no validity in relation to purely intraprovincial transactions and in that respect is ultra vires" (70) In Clark, the Court held that a limitation period in the federal Railway Act, (71) which it acknowledged "was intended to apply to common law negligence actions" (72) against railways, was invalid in respect of such actions because "a limitation provision relating to an action for personal injury caused by a railway cannot be said to be an integral part of federal jurisdiction." (73) It added, however, that the provision could be given a valid interpretation if its application were restricted to statutory causes of action created under the Railway Act: "Although Parliaments intent may not have intended [sic] to restrict the limitation provision to those causes of action specifically created by the statute, section 342(1) is capable of bearing that restricted meaning." (74) In Isen, the Court concluded that federal jurisdiction over pleasure craft under the Canada Shipping Act (75) did not extend to their owners' attempts to secure them to trailers for transport on provincial highways. As a result, the provision in the Act limiting the maximum quantum of damages payable in claims "occurring in direct connexion with the operation of [a] ship" (76) had no application when the harm resulted from the owner's careless deployment of a bungee cord in an effort to secure his boat to a trailer for highway transport. (77)

The Court's approach in these decisions, Clark especially, certainly resembles a typical IJI analysis, albeit in reverse. It's interesting, though, that the Court seems not to have recognized, or to have chosen to label, as IJI what it was doing in these cases. (78) This fact, and the paucity of case law available in support of reciprocal IJI, invites conjecture that the Supreme Court's current apparent preference for reciprocity is based on something other than precedent.

There is a nice symmetry to the notion that IJI works both ways, protecting as appropriate the cores of provincial, as well as those of federal, heads of power from inadvertent interference at the hands of the other order of government. I remain open to persuasion that this is the better approach. But at present, I think it probably isn't, for at least three reasons.

First, it envisages enclaves of provincial subject matter that are absolutely off-limits, even to the inadvertent effects of valid federal legislation and even when that legislation is aimed at addressing and ameliorating some genuine national crisis or emergency. It's true that the courts could mitigate this risk from case to case by gerrymandering descriptions of the cores of the relevant heads of provincial power or by adjusting upward the degree of interference necessary in such instances for the federal law to be deemed inapplicable. But such stratagems, especially if invoked reactively, at the last minute, risk seeming makeshift and even opportunistic.

Second, and more importantly, consider the relationship between IJI and the paramountcy doctrine. In their traditional applications, IJI and paramountcy complement one another, providing, where appropriate, alternative means of protecting federal law and legislative authority. But if IJI were to protect exclusive provincial authority, the two doctrines would operate at cross-purposes, at least in disputes involving both federal and provincial legislation. (79) In such a case, IJI would render the federal law inapplicable; paramountcy, by its very nature, (80) would give the federal law such priority as it required to operate free of interference from the provincial law. In a significant number of cases, therefore, the outcome would turn on which of these doctrines the Court chose to apply, or the order in which it chose to apply them.

The Supreme Court has considered this question of sequence from time to time. In Law Society of British Columbia v Mangat, (81) it held that "[t]he existence of a double aspect to the subject matter of [the relevant federal legislation] favours the application of the paramountcy doctrine rather than the doctrine of [IJI]." (82) But a double aspect will nearly always be involved when valid provincial legislation can plausibly be argued to implicate exclusive federal authority or vice versa; that is just what a double aspect is. If one fully accepted this view, it would be hard to imagine a case in which a court could invoke IJI on behalf of either order of government; resolving the paramountcy issue one way or the other would conclude the conversation. And in COPA, the Court held expressly that the presence of a double aspect was not a complete answer to a claim based on IJI, pointing to previous occasions on which it had used IJI despite a double aspect. (83)

Relying in part on Mangat, (84) the Court in Canadian Western Bank articulated a somewhat more nuanced approach. "[W]hile in theory a consideration of [IJI] is apt for consideration after the pith and substance [validity] analysis," it said, "in practice the absence of prior case law favouring its application to the subject matter at hand will generally justify a court proceeding directly to the consideration of federal paramountcy." (85) In both Ryan Estate (86) and Rogers, (87) the Court quoted this passage with approval; in PHS, it said simply that "before applying the doctrine of [IJI] in a new area, courts should ask whether the constitutional issue can be resolved on some other basis." (88) It was because there were relevant precedents that the Court in Ryan Estate and Rogers considered IJI before, or in lieu of, paramountcy. (89) It did so in COPA without explanation, but there too it had precedent for considering IJI in relation to the relevant head of federal legislative authority, (90) so COPA, too, is consistent with the approach prescribed in Canadian Western Bank.

But consider what happens when we apply this approach when invoking IJI on behalf of provincial authority. As we have seen, there is, at most, a paucity of jurisprudence reading down federal legislation to protect something deemed essential to provincial legislative authority, (91) and even those decisions do not invoke the doctrine of IJI to explain or to justify what they do. This means that there will almost always be an "absence of prior case law favouring [IJI]'s application to the subject matter at hand" (92) when the subject matter at hand is a head of exclusive provincial legislative authority. (93) According to the reasoning in Canadian Western Bank, therefore, at a minimum, it will "generally" be appropriate for a court in such cases to proceed "directly to the consideration of federal paramountcy." (94) On this view, therefore, IJI generally will be meaningfully available to protect provincial authority only where the relevant federal legislation poses no threat to any existing provincial legislation. This seems counterintuitive, to say the least. (95) But the alternative--inverting the sequence of inquiry only when IJI is invoked to protect exclusive provincial legislative authority--seems tendentious and unprincipled, at best.

Finally, the notion that IJI exists to protect provincial as well as federal legislative authority sits uncomfortably with section 91 of the Constitution Act, 1867. "The basic principle of the division of legislative powers in Canada," Pigeon J said in 1976 for a plurality of the Court, "is that all legislative power is federal except in matters over which provincial legislatures are given exclusive authority." (96) The opening words of section 91 subtract "the Classes of Subjects by this Act assigned exclusively to the Legislatures of the Provinces" from the federal orders residual power "to make Laws for the Peace, Order, and good Government of Canada" but add explicitly that "notwithstanding anything in this Act"--despite, therefore, sections 92, 92A, and 93, which prescribe the scope of exclusive provincial legislative authority--"the exclusive Legislative Authority of the Parliament of Canada extends to all Matters coming within the [29] Classes of Subjects next hereinafter enumerated". (97) And according to the closing words of section 91, "any Matter coming within the Classes of Subjects enumerated in this Section shall not be deemed to come within the Class of Matters of a local or private Nature comprised in the Enumeration of the Classes of Subjects by this Act assigned exclusively to the Legislatures of the Provinces." (98) The combined effect of these two phrases is to subtract what comes within the enumerated heads of federal legislative authority from what remains within the enumerated heads of provincial legislative authority. (99) But if the enumeration of exclusive federal authority operates to reduce the scope of exclusive provincial authority, then it follows necessarily that no valid exercise of enumerated federal power can possibly interfere unduly with (what is left of) exclusive provincial power. It follows, in other words, that there cannot be cores of exclusive provincial authority immune from the incidental or even unintended effects of valid federal legislation.

We need to take care not to overstate this point. Subtraction of the federal power over "Marriage and Divorce" (100) from exclusive provincial authority over "The Solemnization of Marriage in the Province," (101) for example, or the plenary federal power of taxation (102) from the more circumscribed provincial taxation power, (103) would neutralize the relevant provincial authority, and subtraction from provincial authority of a robust federal power over "The Regulation of Trade and Commerce" (104) would leave the provinces with not much to do at all. (105) To address these discrepancies, it's been necessary to construe more narrowly the scope of these heads of enumerated federal power, (106) in particular that over trade and commerce, (107) to leave some room for meaningful exercise of the counterpart provincial powers. But subject to these adjustments, the scheme of the distribution of powers in the Constitution Act, 1867 seems clear and appears to leave little, if any, room for cores of exclusive provincial power immune from the inadvertent effects of federal legislation, found upon proper and prudent characterization of the reach of federal authority, to be valid.

Despite recent suggestions to the contrary, the better current view remains that IJI protects exclusive federal, but not exclusive provincial legislative authority. Our next task is to consider what contribution, if any, IJI really makes to the project of Canadian federalism.

C. WHAT GOOD IS IJI?

What practical difference does IJI make to the preservation of Canadian federalism? To begin with, it operates to ensure that the core of exclusive federal legislative authority is as well protected from valid as from invalid provincial legislation. "The rule of the exclusive nature of fields of jurisdiction," the Court said unanimously in Bell 88, "does not depend on a legislative drafting technique." (108) "Nothing," Laskin CJC, writing for the plurality, added in Natural Parents, "accretes to provincial legislative power by the generalization of the language of provincial legislation if it does not constitutionally belong there." (109) As Robin Elliot explained in 2008:
A provincial legislature that decided to enact a generally worded--and,
we can assume for present purposes, perfectly valid--statute that the
provincial government subsequently seeks to apply in a context that
arguably falls within the exclusive jurisdiction of Parliament should
be neither worse not better off than a provincial legislature that
specifies, in a long list of contexts within which its virtually
identical statute is to apply, that very same arguably federal context.
(110)


IJI is the doctrine that preserves this equivalence. It is what consistently secures the exclusivity of exclusive federal legislative authority.

And by so doing, as COPA points out, IJI preserves "Parliament's legislative freedom." (111) The issue there was the impact that restrictions set out in valid provincial agricultural zoning legislation had on Canada's capacity to select locations for aerodromes. "If Parliament wished to override [the relevant provincial law] by way of federal paramountcy," the majority concluded,
it would be forced to establish a legislative conflict with each of the
[decisions of the relevant provincial authorities] regarding aerodromes
.... Parliament would not be free to introduce broad, permissive
legislation, should it so choose (and as it has chosen to do).
Acceptance of this argument would narrow Parliament's legislative
options[.] (112)


IJI excuses the federal order from the need for such legislative busywork or for such vigilance, in respect of those select core matters to which it pertains. In Rogers, the Court again invoked IJI to protect the fruits of a broad permissive federal regime, this time from intrusive municipal requirements. (113)

But if, thanks to IJI, Parliament may safely choose to "introduce broad, permissive legislation" in respect of some core federal matter, another of "Parliament's legislative options", no less worthy of IJI's protection, is electing, in the public interest, to leave some such matter altogether unregulated. (114) Suppose, for example, that Parliament concludes that private ordering, the common law, (115) or even customary Indigenous law suffices to manage the issues and transactions that arise within a given sphere of exclusive federal authority--that legislation of any sort within that sphere would make things worse, not better. But for the doctrine of IJI, neither Parliament nor anyone else could ensure realization of that arrangement. (116) Put differently, IJI also preserves Parliament's freedom not to legislate--its capacity to protect certain matters from legislation altogether. Deciding whether to enact legislation, the Supreme Court observed recently, is also "a valid exercise of parliamentary sovereignty" and "entirely a matter of political choice." (117) Without IJI, the capacity to make that choice would be compromised substantially. (118)

IJI, therefore, contributes distinctively to Canadian federalism; it does real work that otherwise would not, could not, get done. By providing complete, consistent protection to that which is exclusive to federal legislative authority, IJI also protects within that domain Parliament's freedom to legislate, or not to legislate, from unacceptable complications that result, even inadvertently or accidentally, from otherwise permissible provincial legislative activity. When it matters more which order of government does what than whether, for instance, some valid legislation is better than none, IJI is the doctrine of ultimate recourse.

But when, if ever, is this the case? Does the work that IJI does really need doing? We know the Supreme Court's current answer to these questions: not very often at all. It is time that we revisit and appraise the Court's critique of IJI.

IV. THE SUPREME COURT'S CRITIQUE OF IJI

As we saw much earlier, (119) the Supreme Court has offered six reasons for keeping IJI on a very short leash. Each deserves consideration individually.

A. DOES THE PARAMOUNTCY DOCTRINE SUFFICE?

According to the majority in Canadian Western Bank, IJI "would seem as a general rule to be superfluous in that Parliament can always, if it sees fit to do so, make its legislation sufficiently precise to leave those subject to it with no doubt as to the residual or incidental application of provincial legislation." (120) If Parliament isn't sufficiently concerned about the operation of a provincial law to enact legislation ousting it, does the law pose a constitutional problem in need of a judicial solution?

One might have thought that the Court's decision a few years later in COPA had put an end to this line of argument. (121) In that case, which featured no paramount federal legislation, (122) the Court invoked IJI precisely because, in its view, an expectation that Parliament enact such legislation would interfere excessively with "Parliament's legislative freedom." (123) And if the Supreme Court were serious about treating IJI as reciprocal, available to protect provincial as well as federal authority, (124) it would have recognized that the option of enacting paramount legislation is open only to the federal order. (125) But four years later, in Tsilhqot'in, we see once again the suggestion that validity and paramountcy analysis are sufficient and preferable in dealing with division of powers disputes. (126) So here we are.

Concluding that paramountcy suffices to protect federal legislative authority from consequential impairment by valid provincial legislation--telling the federal order, in effect, use it or lose it--has at least three unfortunate consequences.

First, it means that only the federal order itself can protect its exclusive authority from the consequences of inadvertent but inappropriate provincial interference. It's true, of course, that the rest of us may challenge provincial legislation on paramountcy grounds, but we can do so plausibly only when Parliament has already enacted a scheme that might oust such legislation. If we accept (as argued above) that everyone has an interest in maintaining and enforcing the division of powers, (127) no such restriction is defensible. The federal order's view is not the only relevant one about what is exclusive to federal legislative authority or about a provincial law's impact on that authority.

Secondly, burdening Parliament with exclusive responsibility for acting to displace inadvertent incursions on the core of its authority imposes additional (and likely unwelcome) pressure on its legislative agenda. Although Parliament has the constitutional capacity to enact valid paramount legislation more or less whenever it chooses, in practice things are typically more challenging. For one thing, there can be no assurance that any given federal measure will suffice to displace existing unwelcome provincial legislation: that courts will find sufficient incompatibility to render the provincial law inoperative. (128) And there is only so much time in any given legislative session to develop policy and to draft and consider bills; competition for that time among legitimate items of public business can be intense. Expending scarce resources on legislative efforts to supersede inopportune provincial legislation means deferring other public business.

Finally, insisting on the sufficiency of the paramountcy doctrine overlooks the role that IJI traditionally has played in preserving and maximizing Parliament's options for deployment of its own legislative authority and the fact, as mentioned above, that the federal order may sometimes choose, in its own best view of the public interest, to leave unregulated some matters that lie within its authority. (129)

This brings us to "legal vacuums."

B. LEGAL VACUUMS

Among the Court's most frequent recent complaints about IJI has been that its use results in legal vacuums. (130) In this, of course, IJI is hardly unique. Declarations of invalidity--on grounds of federalism, the Charter, section 35, (131) section 96, (132) or the language of enactment--sometimes result in quite capacious legal vacuums, but no one, to my knowledge, has suggested that courts should decline on this account to enforce the Constitution. (133) But IJI, for some reason the Court has yet to articulate, makes the Court selectively uneasy about the prospect of legal vacuums. In PHS, the Court went so far as to suggest that IJI might create legal vacuums that neither order of government is capable of filling. (134)

Surely that's not correct. We have known since 1912 that the Constitution Act, 1867 distributes the entire spectrum of mainstream constitutional authority exhaustively between the federal and provincial orders of government: (135) that "there is no topic that cannot be legislated upon." (136) If one order of government lacks constitutional authority to regulate some matter of public interest, (137) it follows necessarily that the other has such authority. And from this, it follows that any given legal vacuum is, at bottom, discretionary.

What, exactly, is a legal vacuum? There are, of course, many matters of potential public importance that are not, at any given time, subject to governing legislation. But in the absence of such legislation, it is not as though there is no law at all on which to draw for just resolution of such disputes as may arise. (138) At an absolute minimum, all the private law learned and taught in law school--property, contract, tort, restitution, trust and fiduciary law, and the rest of the law of equity--subsists and is adaptable and enforceable (in the common law jurisdictions, at least) as circumstances require. Customary Indigenous law may fill some additional gaps. No doubt these bodies of law will not always be adequate to the circumstances (that is, after all, why we have legislation), but who, in a democracy, gets to decide when they are not?

Exactly. When legal vacuums exist, it is because a competent legislative body has overlooked them, left them for another day, or chosen consciously to leave them unaddressed. All it need ever do to address them is enact the appropriate legislation. (139) When the Court expresses alarm about a legal vacuum, what it is really doing is substituting its judgment for that of the relevant legislature about the desirability of a legislative solution: a role it has traditionally forsworn. (140)

Filling a legal vacuum is not necessarily easy. Legislative time, as noted above, (141) is scarce and coveted. But it is, at a minimum, no more difficult for Parliament to close a legal vacuum than to enact legislation meant to displace a provincial law. If anything, it is less so, for Parliament need not contrive to conflict with any existing provincial legislation.

C. COOPERATIVE FEDERALISM

"[C]onstitutional doctrine," the Court asserts in Canadian Western Bank and repeats in Tsilhqot'in, "must facilitate, not undermine, what this Court has called 'co-operative federalism.'" (142) "Where possible," it adds in PHS, "courts should allow both levels of government to jointly regulate areas that fall within their jurisdiction." (143) This, the Court says, is another reason to constrain the use of IJI.

Whether some particular matter falls within the jurisdiction of both orders of government (such that both may, if they choose, regulate it jointly) is, of course, the very question that IJI exists to help answer. And we know from recent Supreme Court of Canada jurisprudence that cooperative federalism, "an interpretive aid," exists to "foster... cooperation between Parliament and the legislatures within the existing constitutional boundaries" not to extend or constrict either order's actual legislative authority. (144) Presuming concurrent jurisdiction, as the Court's recent IJI jurisprudence appears to do, rigs the contest from the outset, leaving little room for dispassionate discussion of IJI's merits.

And is there really a problem here? Just how inconvenient is IJI to the realization of cooperative federalism? Experience suggests that the two orders of government have been cooperating (when that suited them) quite successfully for quite some time, despite, or irrespective of, occasional judicial resort to IJI. The Supreme Court has already blessed, for instance, arrangements in which the federal order has: (1) given federal effect to local fisheries regulations the provinces have drafted, delegating implementation and enforcement of those regulations back to the provinces; (145) (2) made it a criminal offence for gaming establishments to operate without first obtaining, or without complying with the terms and conditions of, provincially issued gaming licences; (146) and (3) proposed to delegate to a council of participating federal and provincial ministers authority to make regulations governing aspects of nationwide trade in securities that come within federal legislative authority. (147) Both orders of government have considerable scope to delegate legislative authority (just not to each other) (148) and administrative powers (even to one another). (149) And either may elect at will to incorporate by reference, then apply and enforce as its own law, such portions as it chooses of the law of the other, as that law may evolve from time to time. (150) The absolute worst one can say is that IJI sometimes requires creative workarounds in designing mutually desirable cooperative arrangements. It's remarkable that the Court, for all its solicitude on behalf of cooperative federalism, has yet to point to an instance where IJI has frustrated its operation.

Finally, this argument against IJI appears to assume that jurisdictional overlap, or even joint regulation, entails cooperation between the two orders of government and that overlapping regulation is more effective regulation. Not necessarily--when each order of government may legislate as it will, either may elect to cooperate with the other for some appropriate public purpose, but there is no constitutional imperative compelling it to do so (151) or (apart from paramountcy) constraining it if it does not. (152) "One could plausibly argue," Robin Elliot observed in 2008, "that if the courts think it important to encourage co-operation between the two orders of government, they should work to reduce rather than enhance the degree of overlap between federal and provincial jurisdiction." (153) And as Peter Hogg and Rahat Godil observed in that same year, provinces lack the expertise that federal regulators have to design schemes appropriate for the regulation of federal undertakings. (154)

D. THE CENTRALIZING TENDENCY

According to the majority in Canadian Western Bank, "a broad use of the doctrine of [IJI] runs the risk of creating an unintentional centralizing tendency in constitutional interpretation," because IJI has traditionally protected federal, but not provincial, legislative authority from inadvertent incursion. (155) "The 'asymmetrical' application of [IJI]," it says, "is incompatible with the flexibility and co-ordination required by contemporary Canadian federalism." (156)

Again, if the Court were serious in its recent affirmations that IJI works both ways, it would recognize this not as a problem endemic to IJI, but as a historically contingent fact about the cases and situations that, so far, have given rise to IJI jurisprudence. If IJI really does operate reciprocally, all the Court need do to redress any centralizing tendency is set about invoking it on behalf of exclusive provincial authority, so that the jurisprudence can even out. (157) This, of course, would involve an expansion, not a contraction, of IJI's range and relevance.

For the reasons given above, however, my view is still that IJI operates asymmetrically, principally because the text of the Constitution Act, 1867 confirms the exclusivity of federal authority "notwithstanding anything in this Act" (158) and instructs us to subtract the enumerated heads of exclusive federal authority (once we've scoped them appropriately) from the enumeration of exclusive provincial authority. (159) And the Court's current view is that IJI's "natural area of operation is in relation to those heads of legislative authority that confer on Parliament power over enumerated federal things, people, works or undertakings." (160) Substantially fewer enumerated things, people, works, or undertakings lie within the catalogue of exclusive provincial legislative authority. On this understanding, therefore, there is nothing "unintentional" about the "centralizing tendency" that IJI enables; it is, for better or worse, a tendency built into Canada's current constitutional geography. (161)

It is nonetheless a tendency that the Court seems disposed to resist, because, in the Court's view, it is "incompatible with the flexibility and co-ordination required by contemporary Canadian federalism." (162) We have already had occasion to wonder how incompatible IJI is with efforts at federal-provincial coordination. (163) But even if we accept that point, the underlying normative questions--just how much flexibility and coordination does Canadian federalism really require, and why?--still await answers. This brings us to "the dominant tide". (164)

E. THE DOMINANT TIDE

We keep hearing about the dominant tide in Canadian constitutional jurisprudence, which the Court tells us favours a more flexible and permissive federalism. IJI, the Court insists, needs constraint because it complicates this more permissive federalism. (165)

In one sense, of course, the dominant tide in Canadian constitutional jurisprudence is whatever the Court decides it is, from time to time. (166) But this view of things assumes that this more tolerant federalism makes optimal sense today of Canada's constitutional foundations. And is that really so?

I think we can safely say that the proposition isn't obvious. Much earlier, we saw that there would be risks to both orders of government, and perhaps especially to provincial autonomy, from a regime that permitted each to enact whatever it wanted, and that the rest of us have at least an implied constitutional right--assuming we accept that the existing division of powers is at least sufficiently rational to be worthy of enforcement--to invoke constitutional discipline against either if it appears to be overreaching. (167) There is such a thing, then, as too much flexibility, too much permissiveness when it comes to Canadian federalism; relaxing the boundaries too much comes at a cost. And we know the Court knows that because there are limits to its own enthusiasm for permissive federalism. Even at its most ardent, the Court's advocacy for the dominant tide has stopped short of unqualified endorsement. "[N]otwithstanding the Court's promotion of cooperative and flexible federalism," it reminds us more than once, "the constitutional boundaries that underlie the division of powers must be respected. The 'dominant tide' of flexible federalism, however strong its pull may be, cannot sweep designated powers out to sea, nor erode the constitutional balance inherent in the Canadian federal state." (168)

But once we acknowledge that there are and ought to be limits to the pull of the dominant tide, we find that talk of the dominant tide ceases to be useful analytically. Pared to its essence, it is just a rhetorical substitute for the hard work of discerning what matters when, and why. More helpful would be reasoned discussion of why a flexible federalism is desirable (when it is desirable), and what deserves protection despite it, and why (when it is not).

F. POSTULATING ABSTRACT CORES

Finally, there is this concern about IJI's methodology. It is best for me to present it in the Court's own words:
Excessive reliance on the doctrine of [IJI] would create serious
uncertainty. It is based on the attribution to every legislative head
of power of a "core" of indeterminate scope--difficult to define,
except over time by means of judicial interpretations triggered
serendipitously on a case-by-case basis. The requirement to develop an
abstract definition of a "core" is not compatible, generally speaking,
with the tradition of Canadian constitutional interpretation, which
favours an incremental approach. While it is true that the enumerations
in ss. 91 and 92 contain a number of powers that are precise and not
really open to discussion, other powers are far less precise, such as
those relating to the criminal law, trade and commerce and matters of a
local or private nature in a province. Since the time of Confederation,
courts have refrained from trying to define the possible scope of such
powers in advance and for all time:... It was by proceeding with
caution on a case-by-case basis that the courts were gradually able to
define the content of the heads of power of Parliament and the
legislatures, without denying the unavoidable interplay between them,
always having regard to the evolution of the problems for which the
division of legislative powers must now provide solutions. (169)


I confess that I struggle to understand this concern. A minor point to begin with: the core of a given head of power is conceptually distinct from, and generally not coextensive with, the entirety of that power's scope or content. We know, for example, that sections 42-47 of the Indian Act, (170) which make special provisions for the wills of registered "Indians" ordinarily resident on reserve, (171) are valid federal legislation, (172) but from this it does not follow that provincial estates legislation can have no application to the wills of registered Indians ordinarily resident elsewhere, or to Inuit or Metis, (173) to whom that Act does not apply. Likewise, a provision in federal legislation that, considered alone, would be ultra vires may yet be rescued from invalidity if it is "sufficiently integrated" within an otherwise valid federal legislative scheme (174) (or vice versa). (175) Such provisions come within the outer limits of federal legislative authority despite their evident remoteness from the "basic, minimum and unassailable content" (176) that anchors that authority. As the Court itself acknowledged elsewhere in Canadian Western Bank, it is prudent to keep these two concepts, scope and core, distinct. (177) Conclusions about the methodology appropriate to the one may not be as appropriate to the other.

Second, the Court's reasoning in this passage appears to prove too much. If development of "an abstract definition of a 'core'" (178) really is both essential to the conduct of IJI inquiry and incompatible with the tradition of Canadian constitutional interpretation, that defect has also compromised any use of IJI past or present, not just potential future ones. If this circumstance gives the Court sufficient reason to fear the consequences of deploying IJI prospectively (and if it does not, why mention it?), the appropriate response is not merely to curtail IJI's future use, but to abandon the doctrine altogether. That result, itself difficult to reconcile with any sensible notion of exclusive legislative authority, is one the Supreme Court has stopped well short of embracing.

But third, and most importantly, the Court provides neither reasoning nor authority in support of its claim that IJI requires development of "an abstract definition of a 'core'... in advance and for all time". (179) All that IJI requires is that courts be able to tell, in any case where the issue arises, whether any given federal matter qualifies for immunity from potential provincial interference and, if it does, whether a given provincial measure interferes sufficiently with it to trigger the immunity. Cores need have no existence independent of the results of such inquiries; "core" is just a label for the category of matters found after individual consideration to qualify for such protection in respect of a given head of power. There is no obvious reason why our developing sense of what that category contains cannot, nor should not, emerge "over time by means of judicial interpretations triggered serendipitously on a case-by-case basis" or, somewhat more charitably, via the salutary process of reconciling provisional general principles with intuitions and judgments about particular situations that John Rawls has called "reflective equilibrium" (180) and Nelson Goodman "virtuous circularity". (181) Put differently, there is no obvious methodological reason why "an incremental approach" compatible "with the tradition of Canadian constitutional interpretation" (182) cannot yield results in respect of IJI as trustworthy as those it yields in respect of validity or paramountcy.

And if this is so, then it seems unnecessary to stipulate a priori ("in advance and for all time" (183)) that IJI is simply unavailable to protect matters coming within certain heads, or kinds of heads, of federal legislative authority. (184) It may well turn out that different heads of federal power fare differently within IJI inquiry. If so, that needn't be reason for particular concern. If we accept here, as we do elsewhere, that the burden falls on parties asserting IJI to establish it, the risk of "serious uncertainty" (185) seems no greater in IJI inquiry than it is elsewhere in Canadian constitutional jurisprudence. In the absence of a genuine methodological concern, looking and seeing, case by case, seems preferable to making sweeping assumptions or pronouncements at the outset of further inquiry into the reach of IJI. By way of example, the Court was mistaken to say, as it did later in PHS, that IJI "has never been applied to a broad and amorphous area of jurisdiction." (186) In COPA, decided the year before, the Court invoked IJI in support of Parliament's "exclusive jurisdiction to determine the location of aerodromes" (187) Parliament's legislative authority over aerodromes, and over aeronautics more generally, derives, as the Court acknowledged there, from its residual power to make laws for the peace, order, and good government of Canada (POGG) in respect of matters not within exclusive provincial legislative authority. (188) It's hard to imagine a head of power more "broad and amorphous" (189) than POGG. If Canadian federalism can survive ascription of a core to POGG, it is difficult to justify assuming that greater jeopardy would result from entertaining the thought that any given enumerated head of federal authority might prove to have a core, as well.

G. SUMMARY

None of the reasons the Court has offered in its recent decisions, then, suffices to substantiate its disposition to limit IJI's reach. That conclusion gives us reason to doubt the soundness of the constraints the Court has imposed on the doctrine. But from this, no other conclusions follow about IJI's proper reach. The underlying normative inquiry on which such conclusions depend still awaits. How ought we to think about IJI in order that we may trust the results, whatever they are, of subsequent IJI inquiries?

V. RETHINKING IJI

Two questions arise in considering when and how to deploy IJI: what (if anything) warrants its protection; and, where it applies, what (if anything) it renders inapplicable. The first of these questions concerns what comprises the core of exclusive federal legislative authority; the second, the threshold beyond which inadvertent provincial interference with core federal matters becomes impermissible. Let's consider these questions in turn.

A. WHAT DESERVES IJI'S PROTECTION

Precedent, as we saw at the outset, (190) became in Canadian Western Bank the touchstone of contemporary doctrine concerning IJI. In subsequent decisions, the Court has insisted repeatedly on the primacy of precedent in determining what kinds of matters IJI protects. (191) "Predictability, important to the proper functioning of the division of powers, requires", it said in PHS, "recognition of previously established exclusive cores or power."192

This is at least somewhat surprising. Wasn't the whole point of the Court's detailed critique of IJI to correct a propensity it perceived in the previous jurisprudence to deploy IJI too liberally? Strange, then, that its reformed approach would advocate giving pride of place to precisely those matters deemed essential to federal authority during eras preceding its current fondness for more permissive federalism. (193) One might have expected these to be the relics most vulnerable to the dominant tide. It was just such perceived obsolescence that prompted the Court in Tsilbqot'in to overrule a 2006 decision, released while Canadian Western Bank was still under reserve, (194) that had included Indigenous peoples' treaty rights in the core of exclusive federal authority over "Indians, and Lands reserved for the Indians", (1)'" and to omit mention altogether of a 1997 ruling (196) that had placed Aboriginal rights and title within exclusive federal legislative authority. (197) When inconvenient precedent can be cast aside so easily, the Court's professed respect for precedent seems not only incongruous, but selective. (198) Selective reliance on precedent hardly promotes doctrinal coherence or predictability. It cannot provide a trustworthy foundation for limits on the permissible reach of valid provincial legislation.

Neither, I suggest, can the accident of prior judicial attention. The effect of Canadian Western Bank was to discourage prospective resort to IJI except as regards (some of) those grandfathered heads of federal authority in respect of which courts had already approved such resort. But as far as we know, it is a matter of historical happenstance that only certain heads of federal legislative authority attracted litigation and relevant judicial scrutiny before 2007, while others did not. That cannot be sufficient reason for differentiation, especially when, as the majority reasons acknowledge in Canadian Western Bank, IJI "is in principle applicable to all federal and provincial heads of legislative authority". (199)

As we have seen, (200) no methodological complication disqualifies any enumerated head of federal legislative authority from consideration as a potential locus for IJI. This must be so if even Parliaments residual POGG power, from which all provincial legislative authority is subtracted, can have an exclusive core. (201)

In fairness, the Supreme Court has not completely foreclosed recourse to IJI in respect of matters not yet covered by precedent. In Canadian Western Bank, it was careful to say that IJI should "in general" or "largely" be reserved for heads of federal power that have attracted its protection in the past (202) and that omission or postponement of the IJI inquiry is "generally" appropriate in the absence of clear precedent engaging it. (203) Subsequent IJI decisions typically, though not always, have retained this margin for possible innovation. (204) With one arguable exception, however, these decisions have not said clearly when it might be appropriate or permissible to invoke IJI despite the absence of clear precedent invoking it (or to reconsider its appropriateness despite a precedent). Pending such clarification, we cannot know how seriously to take the Court's hesitation to be categorical.

Here, then, is the arguable exception, a portion of which I quoted more briefly above. (205) IJI, the majority held in Canadian Western Bank:
is of limited application and should in general be reserved for
situations already covered by precedent. This means, in practice, that
it will be largely reserved for those heads of power that deal with
federal things, persons or undertakings, or where in the past its
application has been considered absolutely indispensable or necessary
to enable Parliament or a provincial legislature to achieve the purpose
for which exclusive legislative jurisdiction was conferred, as
discerned from the constitutional division of powers as a whole, or
what is absolutely indispensable or necessary to enable an undertaking
to carry out its mandate in what makes it specifically of federal (or
provincial) jurisdiction. (206)


The majority in COPA, relying on this passage, treated "absolute necessity" as the test to use generally in discerning the core of federal authority (207) and said of precedent only that it "will frequently serve as a useful guide to identify the core of a federal head of power." (208) The majority judgment in Rogers, however, which quoted in full this passage from Canadian Western Bank, underlined the words "for situations already covered by precedent" and the words "in the past", (209) suggesting that the subsequent text exists, in its view, only to substantiate the paramount importance of precedent in IJI inquiry. The Court has yet to choose clearly between these contrasting interpretations of its own work. In the meantime, current IJI doctrine remains unsatisfying and unstable.

I think it is time we consider a different approach.

Let's begin by recalling some basics. The corpus of federal legislative authority, the Supreme Court has told us, comprises "powers relating to matters of national importance" (210) or "powers better exercised in relation to the country as a whole." (211) Other constitutional doctrines exist to protect such matters from overt or surreptitious provincial attempts to exercise that authority; we do not need IJI for that. Those other doctrines do not, however, protect such matters from the incidental effects of valid provincial measures--the unintended consequences that may result when provinces do exactly what they are entitled to do. Within broad limits, the constitutional order tolerates this, exposing subjects and creatures of federal legislative authority to the impact of validly exercised provincial authority.

The exceptions are those matters of national importance held to come within exclusive federal legislative authority. These are the matters--the only matters--to which IJI pertains. IJI affords them protection from inadvertent effects of valid provincial legislation equivalent to that available to them independently from advertent (and thetefore invalid) provincial interference. The task, quite simply, is to identify which of these matters are of sufficient national importance to qualify for inclusion within exclusive federal authority and to deserve or require from IJI this special extra protection. This, I suggest, is what courts are really doing in IJI cases, regardless of what they may say they are doing. What we want to know when we consider immunizing something from effects of otherwise legitimate provincial legislative activity is whether its implications for the interests, the values, or the safety of the country as a whole are sufficient to justify such immunization.

This way of framing things has certain advantages. It highlights the irreducibly normative character of the IJI inquiry: the fact that choices involving contested constitutional values are in play in making such determinations. It invites attention to the conclusions--or, in the worst case, the unarticulated assumptions--that underpin such choices. In both these ways, it contributes to the transparency of the analysis and leaves room for IJI law to adapt as necessary in response to significant changes in national circumstances.

The answer to this question of sufficient national importance will not routinely be affirmative (and to be clear, no federal matter that fails the test of sufficient national importance seems to me to deserve IJI's protection). There are provinces in Canada for good reasons; they are very different places, and one size most assuredly does not fit all. As the Court recognized in Spraytech, local government is "closest to the citizens affected and thus most responsive to their needs, to local distinctiveness, and to population diversity." (212) Provinces are able to engage on a local scale, where the stakes are generally somewhat lower, in instructive legislative innovation that, untested, might well seem too risky for deployment nationwide. We do not want to limit the reach of provincial authority without sufficient reason. Anyone seeking to invoke IJI should bear the burden of demonstrating entitlement to its protection.

But when a matter is held to be of sufficient national importance, we have reason to be cautious about the impact of valid provincial legislation upon it. The problem is not that provincial legislatures and governments are inattentive to the national interest, but that the Constitution itself precludes them from enacting legislation that seeks realization of national objectives. Virtually every head of enumerated provincial authority contains such terms as "in the Province," "of the Province," or "Provincial" to emphasize the limited territorial reach, and the necessarily local intendment, of permissible provincial legislation. (213) Any provincial attempt--no matter how conscientious--to regulate matters whose range or importance is deemed to transcend the province is going to be impermissible: invalid. Valid provincial legislation, therefore, is necessarily about something other than matters of national importance. Accordingly, any consequences such legislation has for matters of sufficient national importance are, of necessity, inadvertent: fortuitous. The province can have no legitimate interest in giving effect to them in respect of such matters. (These are, remember, consequences that would invalidate the relevant statutory provisions if the courts believed the province had wrought them intentionally.) And precisely because such consequences are (deemed or assumed to be) altogether free of provincial forethought, the nature and magnitude of their impact on such matters is unpredictable and potentially quite detrimental.

When matters of national importance are beyond provincial authority, careful attention to them while designing or implementing regulatory schemes is, quite excusably, beyond the range of provincial expertise. But that is of little comfort to those implicated in the matter of national importance. Hogg and Godil have made this point well:
Provincial regulators will not have thought about the impact of their
laws on federal undertakings, and if they had given the matter any
thought they would lack the expertise that the federal regulators
possess by virtue of being the primary regulator of that business.
Provincial laws will inevitably have unintended consequences for
federal undertakings. Narrowing the doctrine of [IJI], therefore, not
only enhances the exposure or federal undertakings to double
regulation, it risks the infliction of collateral damage from
provincial laws of general application that were not directed at them.
(214)


The question still unanswered, of course, is how to measure sufficiency of national importance: what must be true of a matter for it to warrant IJI protection. I'll offer my own thoughts about this, but I don't pretend that they are definitive. My real aim in this article is to give the IJI inquiry its proper focus, in order that courts in future IJI cases reach more defensible, more compelling conclusions. Constitutional adjudication is an exercise in incremental national self-definition: an ongoing dialogue involving courts, both orders of government, and other interested public and private parties about what makes us who we are, and why. Reliable judgments about sufficient national importance engage, within the framework of our written constitution, deeper normative questions such as those to which Richard Simeon called our attention 35 years ago: questions about the relative significance of the different kinds of communities within the federation (the national, the local, and the linguistic); (215) about "which level can most efficiently and effectively carry out any given responsibility of contemporary government"; and about the proper locus for democratic accountability in respect of any given governmental task. (216) These are proper subjects

for a broader national conversation, the fruits of which are certain to refine the criteria I propose for use in recognizing sufficient national importance. Still, here is my proposal, as a first approximation.

A given federal matter will meet the test of sufficient national importance, I suggest, when we recognize that any government husbandry of that matter requires single-source coordination nationwide, even at the expense of otherwise perfectly legitimate local or regional preferences or concerns. This will be so in either of two kinds of circumstances. The first is where nothing but single-source coordination can realize the purpose underlying the relevant grant of federal legislative authority. Determination, for example, that the parents of Confederation thought it essential to insulate a matter from vulnerability to local preferences or idiosyncrasies (217) would serve this purpose sufficiently in the absence of good reason to doubt today the soundness of their judgment. The second is where the Constitution permits single-source coordination of the relevant matter (which, of course, is by no means everywhere) and the national public interest requires it. Examples of this latter category may include situations in which the benefits of nationwide standardization clearly outweigh the benefits of local customization and innovation or those in which substantial numbers of Canadians, irrespective of province of residence or default system of law, share essential interests best served by pooled resources. A convenient sign that a matter requires single-source coordination for either of these reasons, and therefore comes within exclusive federal legislative authority, is that we deem it necessary, despite the bijural character of Canada's legal system (comprising both common law and civil law jurisdictions), for the legal rules that govern the matter to be substantially consistent, if not uniform, throughout the whole of Canada. This is so, for example, when the matter is the subject of a recognized body of federal common law: maritime negligence law (218) and the law of Aboriginal title (219) are perhaps the most familiar recent examples. (220)

It will be interesting to see how much difference adoption of this somewhat different understanding of IJI and of exclusivity makes to the outcomes of actual IJI cases. Perhaps the Supreme Court's current penchant for permissive federalism--the notion that letting someone do something is presumptively more important than taking care about who does what--will prove defensible even within this framework. But if I am right, its successful defence will require better arguments than those the Court has so far offered about redundancy, legal vacuums, dominant tides, putative methodological impediments, and hypothetical barriers to cooperative federalism. It will require--as will any persuasive alternative conception--reasoned elaboration of at least some of what is required for Canada authentically to be, and remain, Canada. Acknowledging this more openly and addressing more thematically the issue of sufficient national importance should improve the clarity of the conversation about IJI and the soundness and persuasiveness of its subsequent results.

B. WHAT IJI PROTECTS AGAINST

But identifying in this new way the circumstances apt for consideration of IJI is only half the challenge. What remains is to ponder the threshold beyond which a provincial statute's inadvertent effects on exclusive federal authority--on Parliament's legislative freedom--cease to be acceptable and activate IJI. Has the Court been correct to insist since Canadian Western Bank that only an "adverse consequence" (221) for exclusive federal authority can trigger IJI? If so, what counts as an adverse consequence?

One way of approaching these questions is to compare the Court's decisions in Ordon Estate (1998) (222) and Ryan Estate (2013). (223) Both decisions consider the impact of provincial legislation on maritime negligence law, a field that encompasses, among other things, "the range of possible claimants, the scope of available damages, and the availability of a regime of apportionment of liability according to fault". (224) Both acknowledge that maritime negligence law lies within the core of exclusive federal authority over navigation and shipping (225) and must remain uniform nationwide. (226)

At issue in Ordon Estate was a provision in provincial fatal accidents legislation that included siblings among those authorized to sue for loss of care, guidance, and companionship. The Court deemed the provision constitutionally inapplicable to a maritime negligence action because it had "the effect of regulating indirectly an issue of maritime negligence law" (227) or, alternatively, "the effect of altering existing rules of federal maritime negligence law." (228) Effects such as these, in the Court's view, sufficed to trigger IJI.

At issue in Ryan Estate was a provision in provincial workers' compensation legislation prohibiting civil actions in respect of injuries incurred in the course of employment, requiring instead that compensation for them be sought, if at all, exclusively pursuant to that legislation. This provision too, the Court said "has the effect of regulating a maritime negligence law issue." (229) Even so, it declined to declare the provision inapplicable, because "this level of intrusion into the federal power is insufficient to trigger [IJI]"; (230) "it neither alters the uniformity of Canadian maritime law nor restricts Parliament's ability to determine who may possess a cause of action under" maritime negligence law. (231) If legislation depriving certain individuals altogether of rights of action under maritime negligence law does not suffice to impair exclusive federal authority over such law, one doubts that the legislation considered in Ordon Estate, which only increased the range of litigants eligible under that law, would do so either.

So if substituting a mandatory provincial compensation scheme for a right of civil action anchored in exclusive federal authority does not impair the exclusive core, then what does? In COPA, the Court held that valid provincial land use planning restrictions impaired exclusive federal authority over aeronautics because they impeded a permissive federal scheme about aerodrome locations: a scheme, the Court took pains to show, that did not purport to be exhaustive. (232) "Applying these provincial provisions", the Court in Marcotte said of COPA, "would force Parliament to pass legislation to countermand the provincial rules, failing which the activity could not occur at all." (233) In Rogers, the Court reached much the same conclusion about the impact of municipal restrictions on a permissive federal scheme for siting cellphone towers. (234)

But what differentiates these two scenarios from the one at issue in Ryan Estate} There too, we find juxtaposed a provincial restriction (an outright prohibition on recourse to civil action for work-related injuries) and a permissive federal scheme (creating a cause of civil action where fatality results from maritime negligence). There too, the relevant activity, suing for negligence, "could not occur at all" unless "Parliament... pass[ed] legislation to countermand the provincial rules." (235) It is true that the provincial scheme in Ryan Estate offered recourse to compensation from an alternative source according to an alternative measure. But would the result in COPA or in Rogers have been different--should it have been different--if the relevant provincial schemes there had offered the alternative of erecting something else (an advertising billboard, perhaps) on the relevant site, or an alternative location of the provinces choosing for the aerodrome or the cellphone tower?

Once again, the recent jurisprudence seems neither consistent nor especially illuminating. How might we achieve greater cogency?

The argument I have made so far suggests two fruitful points of anchorage. First, IJI exists to accompany and supplement the ultra vires doctrine. The ultra vires doctrine shields core federal matters altogether from the advertent effects of invalid provincial legislation; IJI affords those same matters equivalent protection from the inadvertent effects of valid provincial legislation. (236) And second, a helpful test for determining what is exclusive to federal legislative authority is to ask which matters are of sufficient national importance to warrant such protection from provincial legislation. (237)

If equivalent protection of core federal matters is our aim, our approach in identifying a trigger threshold for IJI needs to be comparative. We need, in other words, to pay attention to the kinds of effects on exclusive federal authority that, when fully intended, render provisions in provincial law invalid. These are the kinds of effects from which IJI too provides protection. Within the IJI inquiry, the Court has concluded, and now assumes, that the province had no intention of achieving such effects. The relevant question during such inquiry, therefore, is whether such effects would suffice to invalidate the provincial measure if the Court had concluded that they were deliberate: that the province had set out to regulate those federal matters in that way. If the answer is yes, then equivalency requires that IJI apply and immunize the relevant core federal matters from the measure. If the answer is no, then this will not be a case for IJI because, if used, IJI would then be affording the core federal matters greater protection from inadvertent than from advertent effects of provincial law. And that is not the point of IJI.

An approach based on equivalency does not, however, turn on the presence or absence of "adverse consequence." (238) A provincial law may well be invalid despite having welcome (not adverse) consequences for those within federal authority to whom it purports to apply. (Provincial legislation that dispensed Indian status on terms decidedly more generous than those in any federal legislation, for example, would still be invalid; matters relating to Indian status and its necessary incidents lie within exclusive federal authority.) (239) And what would it mean to ask, in the course of a validity inquiry (as COPA instructs us to do in the course of IJI inquiries), whether the impugned provincial measure "seriously or significantly trammels the federal power"? (240) The permissible reach of federal power remains the same regardless of any provincial activity, valid or invalid. Any provincial measure that has the effect of regulating a core federal matter compromises Parliament's freedom to leave that matter unregulated. And any provincial measure that affects a core federal matter, whether trivially or substantially, can suffice (or not) to compel a federal legislative response, depending on the policy orientation of the government of the day (and that, we know, cannot be constitutionally dispositive). Instead, courts ask whether the impugned provincial legislation purports to do something that only the federal order has constitutional authority to do. The relevant counterpart question when considering IJI is whether the valid provincial law has (inadvertent) effects that only the federal order may validly seek to achieve. Either the relevant effect is within exclusive federal legislative authority to achieve, or it is not.

How then are we to identify those effects on core federal matters that only the federal order may validly achieve? I suggested above that the domain of exclusive federal authority include all those matters, but only those matters, that have sufficient national importance to deserve or require immunity from effects of provincial legislative activity. Perhaps there is, or ought to be, a correlation between the comprehensiveness of the immunity that the Constitution affords to core federal matters and the stringency of the criteria we adopt to measure sufficiency of national importance. It makes intuitive sense to suppose, for example, that federal matters found to be of surpassing national importance need and deserve protection from a wider range of effects of provincial legislation (valid or invalid) than federal matters that meet only a lesser (though still, we conclude, sufficient) standard of national importance. The choice, on this view, is between providing maximal protection to a narrower range of federal matters or more relaxed, but still meaningful, protection to a wider range of such matters.

It is premature, I think, to suggest which of these options is preferable. The important thing is that the two inquiries--about the appropriate threshold of national importance and about the kinds of effects wrought by provincial legislation from which core federal matters must be immune--take place with reference to one another. Given the significance and difficulty, conceptual and practical, of these issues, a coordinated inquiry seems most likely to yield perspicuity.

VI. SOME FINAL THOUGHTS

The Constitution of Canada has distributed exclusive legislative authority to the two orders of government not in the only possible way, but in a way that attempts to capture, rationally and sensibly, which kinds of matters require attention from a national perspective and which are dealt with better locally. If we accept the fundamental rationality of this distribution, we have good reason to seek to confine each order of government to its own assigned lane. Each is best positioned, by reason of the perspective and expertise it alone has, to gauge and to anticipate the impact of the relevant legislative options on matters within its authority. Neither has the same capacity to manage the other's choices and responsibilities. Ignoring or erasing the boundary between federal and provincial authority would also risk, at the one extreme, making the federal order all but irrelevant and, at the other, leaving the provinces all but impotent.

The Constitution, therefore, I submit, protects matters of sufficient national importance even from conscientious, well-meant provincial attempts to address them. IJI, I have argued, protects these same matters from inadvertent, unplanned effects of valid provincial legislation when the province could not seek overtly to achieve those effects. Its function is to preserve the exclusivity of federal legislative authority over matters whose significance to the country as a whole is sufficient to require exclusivity. We shall want to consider carefully just which matters those are, but unless we conclude on inquiry that there really are no such matters (in which case, do we really need--or have--a federation?), IJI remains essential within Canadian federalism.

This said, we sometimes pay a price for IJI. When your favourite provincial government enacts attractive, valid legislation that can address a problem your least favourite federal government prefers to ignore, it is inconvenient and frustrating when courts declare the legislation inapplicable. It was because of IJI, for example, that Bell Canada employees found themselves with less substantial occupational health and safety protection than their private sector counterparts. (241) And it was because of IJI that the siblings of those killed or injured in boating accidents had no rights of action against those responsible. (242) Unsympathetic outcomes such as these can make one regret IJI's availability. (243)

In this, of course, IJI is not unique among surviving features of the 1867 Constitution. Many people were disappointed to learn that neither uniform nationwide securities regulation (244) nor robust nationwide legislation on assisted human reproduction (245) was constitutionally permissible, or that constitutional amendment was required to reform the Senate. (246) And one can imagine the reaction of the people of Manitoba to the news that virtually all the legislation enacted in that province was invalid because it was unilingual. (247) Yet those results have not prompted doubts about the ongoing usefulness or propriety of the constitutional doctrines that led to them. Convenience and efficiency are not paramount among federalism's virtues. It is peculiar that IJI, which most often yields consequences much less arresting than any of these, has become the exclusive target for judicial displeasure about the fuss that is federalism.

IJI, like criminal law and the rest of traditional Canadian constitutional law, operates as a hedge against some unwelcome outcomes--but not as an assurance of welcome ones. It is one tool among many in the arsenal of advocacy: a hammer, if you will, when not everything is a nail. Like other tools, it can do harm when misused; it doesn't care who uses it or how. But there are times when no other tool will do. To maximize its utility, we need to learn to recognize these occasions reliably. My argument is that development of this skill requires clearer thinking and better reasoning than the courts have yet devoted to it. In particular, it requires that we pay much closer thematic attention to the question of sufficient national importance, developing clearer criteria, and identifying with proper care the matters that meet this standard. To be fruitful, such inquiry must test candidate criteria with reference to the practical consequences of their application.

If our primary interest is in achieving socially desirable outcomes, we shall also want to learn how and when to use each of the various other tools available in our metaphorical public interest toolbox. These include, no doubt among others, public policy formation, government relations, well informed and strategized political activity, and the rest of what we know as public law. In situations of antecedent constitutional uncertainty--which are frequent in Canada--it will sometimes be prudent to focus efforts at social reform on both mainstream orders of government. But to acknowledge all of this does not require that we overlook or undervalue IJI. No sensible carpenter would disdain her hammer merely because it's a poor substitute for a saw.

KERRY WILKINS ([dagger])

([dagger]) Adjunct professor, Faculty of Law, University of Toronto. Sincere thanks to Carol Rogerson, Joshua Nichols, and the anonymous reviewers and editorial staff at the UBC Law Review for perceptive comments and helpful suggestions on earlier drafts of this article, and to Kent McNeil, Ashley McKenzie, Dana McLean, Diane McMurray, and Constance Marlatt for their suggestions and contributions to a different project that became this one.

(1) Canadian Western Bank v Alberta, 2007 SCC 22 [Canadian Western Bank].

(2) Ibid at para 38.

(3) Ordon Estate v Grail, [1998] 3 SCR 437 at para 83, 166 DLR (4th) 193 [Ordon Estate].

(4) Canadian Western Bank, supra note 1 at para 67. See also ibid at para 42.

(5) Ibid at para 77, quoted at greater length in the text accompanying note 206, below.

(6) Ibid at para 78, quoted at greater length in the text accompanying note 85, below. See notes 83-90.

(7) Ibid at para 48.

(8) See ibid at paras 118-23. See also Robin Elliot, "Interjurisdictional Immunity after Canadian Western Bank and Lafarge Canada Inc.: The Supreme Court Muddies the Doctrinal Waters--Again" (2008) 43 SCLR (2d) 433 at 469, n 157 (arguing that Canadian Western Bank is best understood as a case about the validity of a provision in Alberta legislation aimed specifically at deposit-taking institutions, not a case about the applicability to banks of otherwise valid general provincial insurance legislation).

(9) The decision attracted considerable academic commentary at the time. No fewer than three articles in the 2008 volumes of the Supreme Court Law Review dealt in depth with this development. See John G Furey, "Interjurisdictional Immunity: The Pendulum Has Swung" (2008) 42 SCLR (2d) 597; Peter W Hogg & Rahat Godil, "Narrowing Interjurisdictional Immunity" (2008) 42 SCLR (2d) 623; Elliot, supra note 8. I particularly recommend the Elliot.

(10) See e.g. British Columbia (AG) v Lafarge, 2007 SCC 23 [Lafarge] (released concurrently with Canadian Western Bank, supra note 1); Canada (AG) v PHS Community Services Society, 2011 SCC 44 [PHS]; Marine Services International Ltd v Ryan Estate, 2013 SCC 44 [Ryan Estate]; Tsilhqot'in Nation v British Columbia, 2014 SCC 44 [Tsilhqot'in]; Bank of Montreal v Marcotte, 2014 SCC 55 [Marcotte].

(11) See e.g. PHS, supra note 10 at paras 60-65; Ryan Estate, supra note 10 at paras 49-50; Tsilhqot'in, supra note 10 at paras 144-49; Marcotte, supra note 10 at paras 63-64.

(12) See Quebec (AG) v Canadian Owners and Pilots Association, 2010 SCC 39 at para 44 [COPA]; Rogers Communications Inc v Chateauguay (City), 2016 SCC 23 at paras 60-61 [Rogers]. See also supra note 11.

(13) To be fair, there were some earlier harbingers. Chief Justice Dickson, in particular, was no fan of IJI and felt free to say so. See e.g. Ontario (AG) v OPSEU, [1987] 2 SCR 2 at 17-19, 41 DLR (4th) 1, Dickson CJC, concurring [OPSEU]; General Motors of Canada Ltd v City National Leasing, [1989] 1 SCR 641 at 669, 58 DLR (4th) 255 [General Motors].

(14) See e.g. Canadian Western Bank, supra note 1 at para 77, quoted in the text accompanying note 5, above; Ryan Estate, supra note 10 at paras 49-50; Rogers, supra note 12 at paras 61-63. But not all precedents, it seems, are created equal. See Tsilhqot'in, supra note 10 at para 150; Grassy Narrows First Nation v Ontario (Natural Resources), 2014 SCC 48 at para 53 (rejecting earlier Supreme Court authority that had extended IJI protection to Indigenous peoples' existing Aboriginal and treaty rights). See also notes 192-198 and accompanying text.

(15) See Canadian Western Bank, supra note 1 at para 46; Tsilhqot'in, supra note 10 at para 148.

(16) Canadian Western Bank, supra note 1 at para 44; Lafarge, supra note 10 at para 4; PHS, supra note 10 at paras 64, 69; Tsilhqot'in, supra note 10 at para 147.

(17) See PHS, supra note 10 at para 63; Tsilhqot'in, supra note 10 at paras 148-49. See also Canadian Western Bank, supra note 1 at para 24.

(18) Canadian Western Bank, supra note 1 at para 45.

(19) This proposition originated with Dickson CJC. See OPSEU, supra note 13 at 19. See also Canadian Western Bank, supra note 1 at paras 36-37; Lafarge, supra note 10 at para 4; PHS, supra note 10 at para 62; Ryan Estate, supra note 10 at para 50; Tsilhqot'in, supra note 10 at para 148; Rogers, supra note 12 at para 60.

(20) Canadian Western Bank, supra note 1 at para 43.

(21) Reference re Secession of Quebec, [1998] 2 SCR 217, 161 DLR (4th) 385 [Quebec Secession Reference].

(22) Ibid at para 49.

(23) Ibid at para 55.

(24) Ibid at para 57.

(25) See Canadian Western Bank, supra note 1 at para 21.

(26) Quebec Secession Reference, supra note 21 at para 56.

(27) Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act (UK), 1982, c 11 [Charter].

(28) Constitution Act, 1982, s 35, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.

(29) Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, s 96, reprinted in RSC 1985, Appendix II, No 5 ("[t]he Governor General shall appoint the Judges of the Superior, District, and County Courts in each Province, except those of the Courts of Probate in Nova Scotia and New Brunswick", s 96). See e.g. McEvoy v New Brunswick (AG), [1983] 1 SCR 704, 148 DLR (3d) 25 (neither order of government, nor both acting together, may deprive provincial superior courts altogether of jutisdiction over indictable offences or transform an inferior court into a superior court); McMillan Bloedel v Simpson, [1995] 4 SCR 725, 130 DLR (4th) 385 (section 96 protects a guaranteed core of superior court jurisdiction from both federal and provincial legislative interference).

(30) Constitution Act, 1867, supra note 29, s 121 ("[a]ll Articles of the Growth, Produce, or Manufacture of any one of the Provinces shall, from and after the Union, be admitted free into each of the other Provinces", s 121). See e.g. R v Comeau, 2018 SCC 15 [Comeau] (section 121 "prohibits laws [whether federal or provincial] that in essence and purpose restrict trade across provincial boundaries" at para 106). See generally Comeau, ibid at paras 106-14.

(31) A passage in the majority judgment in Canadian Western Bank suggests, somewhat surprisingly, that the Court is willing to run this risk: "[i]n the absence of conflicting enactments of the other level of government, the Court should avoid blocking the application of measures which are taken to be enacted in furtherance of the public interest": supra note 1, at para 37. Taken seriously, this suggests that the only constraints on the range of federal or provincial authority should be bad faith--enacting something without regard for the public interest--and paramountcy. This could very well be the provinces' worst nightmare. See note 32.

(32) The risks that concurrent jurisdiction poses to provincial autonomy have been a recurring theme in Bruce Ryder's constitutional scholarship. See Bruce Ryder, "Equal Autonomy in Canadian Federalism: The Continuing Search for Balance in the Interpretation of the Division of Powers" (2011) 54 SCLR (2d) 565 at 594-95 [Ryder, "Equal Autonomy" footnotes omitted]:
Because the rule of federal paramountcy renders conflicting provincial
laws inoperative, areas subject to concurrent powers are in fact areas
in which Parliament is ultimately supreme and the provincial
legislatures are subordinate. The provinces have only a conditional
autonomy in areas of de jure or de facto concurrent jurisdiction.
Rather than exercising guaranteed, exclusive jurisdicrion, they are put
in the position of supplicants to the federal government.... If the
provincial pursuit of distinct policies in the growing areas of shared
jurisdiction is conditional upon federal consent or forbearance, the
provinces cannot be confident that their autonomy will be secured in
the future.


In an earlier article, Ryder observed that "[i]f overlapping powers give rise to spheres of hierarchical relations, then an authentic federalism of equals requires that overlapping powers be eliminated": Bruce Ryder, "The Demise and Rise of the Classical Paradigm in Canadian Federalism: Promoting Autonomy for the Provinces and First Nations" (1991) 36 McGill LJ 308 at 345-46. CfWilliam Lederman, "The Balanced Interpretation of the Federal Distribution of Legislative Powers in Canada" in William Lederman, ed, Continuing Canadian Constitutional Dilemmas (Toronro: Butterworths, 1981) 266 at 278.

(33) In Bell Canada v Quebec (Commission de la Sante et de la Securite du Travail), [1988] 1 SCR 749, 51 DLR (4th) 161 [Bell 88 cited to SCR], a unanimous Supreme Court warned of "the risk that these two fields of exclusive powers [federal and provincial] will be combined into a single more or less concurrent field of powers governed solely by the rule of paramountcy of federal legislation. Nothing could be more directly contrary to the principle of federalism underlying the Canadian Constitution": ibid at 766. See also Comeau, supra note 30 ("the federalism principle reminds us of the careful and complex balance of interests captured in constitutional texts. An interpretation that disregards regional autonomy is as problematic as an interpretation that underestimates the scope of the federal government's jurisdiction" at para 82).

(34) See e.g. Reference re Remuneration of Judges of the Provincial Court (PEI), [1997] 3 SCR 3, 150 DLR (4th) 577 ("[t]he institutional independence of the courts emerges from the logic of federalism, which requires an impartial arbiter to settle jurisdictional disputes between the federal and provincial orders of government" at para 124).

(35) Schneider v The Queen, [1982] 2 SCR 112 at 138, 139 DLR (3d) 417, Dickson J (as he then was) [Schneider].

(36) OPSEU, supra note 13 at 19, Dickson CJC, concurring.

(37) Kitkatla Band v British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31 at paras 72-73 [Kitkatla].

(38) See R v Morris, 2006 SCC 59 [Morris].

(39) OPSEU, supra note 13 at 19, quoted with approval in Kitkatla, supra note 37 at para 72.

(40) Nova Scotia (AG) v Canada (AG) (1950), [1951] SCR 31 at 34, [1950] 4 DLR 369, Rinfret CJC [Nova Scotia Interdelegation]. The Chief Justice added that "[t]he country is entitled to insist that legislation adopted under section 91 should be passed exclusively by the Parliament of Canada in the same way as the people of each Province are entitled to insist that legislation concerning the matters enumerated in section 92 should come exclusively from their respective Legislatures": ibid. And according to Taschereau J, "[i]t is a well settled proposition of law that jurisdiction cannot be conferred by consent": ibid at 40.

(41) There have, of course, been a few amendments to the federal and provincial catalogues of powers over the years. See e.g. Constitution Act, 1940, 3-4 Geo VI, c 36 (UK), s 1, reprinted in RSC 1985, Appendix II, No 28 (adding subsection 2A (unemployment insurance) to section 91 of the Constitution Act, 1867, supra note 29); Constitution Act, 1964, 12-13 Eliz II, c 73 (UK), s 1, reprinted in RSC 1985, Appendix II, No 5, repealing and replacing British North America Act, 1951, 14-15 Geo VI, c 32 (UK), s 1 (adding section 94A (old age pensions and supplementary benefits) to the Constitution Act, 1867); and Constitution Act, 1982, supra note 28, ss 50-51 (adding section 92A (provincial legislative authority in relation to natural resources) to the Constitution Act, 1867).

(42) There was some regional variation in the terms of union in respect of legislative authority over education. See Constitution Act, 1867, supra note 29, s 93; Manitoba Act, 1870, 33 Vict c 3, s 22, reprinted in RSC 1985, Appendix II, No 8; Alberta Act, 1905, 4-5 Edw VII, c 3, s 17, reprinted in RSC 1985, Appendix II, No 20; Saskatchewan Act, 1905, 4-5 Edw VII, c 42, s 17, reprinted in RSC 1985, Appendix II, No 21; Constitutional Amendment, 1987 (Newfoundland), SI/88-11 (amending Term 17 of the Terms of Union of Newfoundland with Canada). But these variations are not germane to the present argument.

(43) In a classic 1983 article, Richard Simeon identified three frameworks we might productively use to appraise any given federal scheme or proposal: the way in which it identifies and individuates political communities; how well it equips those in power to satisfy citizens' needs; and what conception of democracy it promotes. See Richard Simeon, "Criteria for Choice in Federal Systems" (1983) 8 Queen's LJ 131.

(44) Alberta (AG) v Moloney, 2015 SCC 51 at para 14 [Moloney]. Cf Canadian Western Bank, supra note 1 ("[e]ach head of power was assigned to the level of government best placed to exercise the power" at para 22).

(45) Moloney, supra note 44 at para 14.

(46) See Albert S Abel, "The Neglected Logic of 91 and 92" (1969) 19 UTLJ 487 (the theme of federal authority "is a bestowal on the Dominion of responsibilities which have as their characterizing trait the management of the economy" together with a handful of powers that "represent 'special situations'--defence and relations with persons outside the political community, to which a common attitude and a central direction are presuppositions of even loosely knit alliances" and "historical residues of the Quebec Constitutional Act's introduction of English criminal law"; on the other hand, "it was the patterns, values and institutions of everyday community contact that were indicated as the legitimate domain of the provinces" at 500-01 [citations omitted]).

(47) See John T Saywell, The Lawmakers: Judicial Power and the Shaping of Canadian Federalism (Toronto: University of Toronto Press for the Osgoode Society, 2002) ("[t]he federal government was endowed with the capacity, the institutions, and the resources to develop and manage a national economy and ensure the expansion and viability of the new state.... The provinces retained their responsibility for the social, cultural, and, to some extent, business life of the community" at 12-14).

(48) Abel, supra note 46 at 504.

(49) Saywell, supra note 47 at 12. Put differently, such terms perforce provided protection for both the rights of linguistic minorities and an essential measure of regional autonomy.

(50) See Thorson v Canada (AG), [1975] 1 SCR 138 at 151, 43 DLR (3d) 1, quoted with approval in Canada (AG) v Law Society of British Columbia, [1982] 2 SCR 307 at 326, 137 DLR (3d) 1 [LSBC]:
The question of the constitutionality of legislation has in this
country always been a justiciable question. Any attempt by Parliament
or a Legislature to fix conditions precedent, as by way of requiring
consent of some public officer or authority, to the determination of an
issue of constitutionality of legislation cannot foreclose the Courts
merely because the conditions remain unsatisfied[.]


(51) A famous example concerns provincial taxes on banks. Section 92.2 of the Constitution Act, 1867 empowers provinces to impose direct taxes "within the Province in order to the raising of a Revenue for Provincial Purposes" and sections 91.15-16 include "Banking" and "Savings Banks" within exclusive federal legislative authority: Constitution Act, 1867, supra note 29. The Privy Council held that a direct provincial tax aimed principally at banks was a perfectly acceptable way for a province to raise revenue. See Bank of Toronto v Lambe (1887), 12 AC 575, [1887] UKPC 29 (BAILII) [Lambe]. But a subsequent provincial tax on banks, the effect of which would have been to render their continued operation within the province financially prohibitive, was legislation in relation to banking, and therefore invalid. See Reference Re Alberta Statutes: The Bank Taxation Act; The Credit of Alberta Regulation Act; and the Accurate News and Information Act, [1938] 4 DLR 433, [1939] AC 117 (PC) [Alberta Taxation Reference].

(52) See e.g. Madden v Nelson & Fort Sheppard Railway Co, [1899] AC 626 at 627-28 (PC), 12 CRAC 224; Saskatchewan (AG) v Canada (AG) (1948), [1949] 2 DLR 145 at 151, [1949] AC 110 (PC).

(53) See originally Tennant v Union Bank of Canada (1893), [1894] AC 31, [1893] UKPC 53 (BAILII).

(54) Consider, for example, the contrasting views of the majority and minority judgments in the trilogy of recent decisions about the paramountcy doctrine and federal bankruptcy legislation. See Moloney, supra note 44; 407 ETR Concession Company v Canada (Superintendent of Bankruptcy), 2015 SCC 52; Saskatchewan (AG) v Lemare Lake Logging Ltd, 2015 SCC 53.

(55) See notes 187-188 and accompanying text.

(56) See Constitution Act, 1867, supra note 29, s 91. See e.g. Canadian Western Bank, supra note 1 at para 41.

(57) In the next section, I consider wherher IJI confers comparable protection from unacceptable federal interference on certain matters reserved exclusively to provincial authority. See notes 63-107 and accompanying text.

(58) See Canadian Western Bank, supra note 1 at para 48, quoted in the text accompanying note 7, above. Impairment, the Court has said, "requires a significant or serious intrusion on the exercise of the federal power. It need not paralyze it, but it must be serious": COPA, supra note 12 at para 45. On this view, IJI does not preclude the application of valid provincial measures that merely affect, or even have the effect of regulating, matters within exclusive federal legislative authority unless the impact amounts to impairment. See Ryan Estate, supra note 10 at paras 62-64.

(59) Bell 88, supra note 33 at 762. See also Commission de transport de la Communaute urbaine de Quebec v National Battlefields Commission, [1990] 2 SCR 838 at 853, 74 DLR (4th) 23 [National Battlefields Commission].

(60) National Battlefields Commission, supra note 59 at 853.

(61) "In short, a legislative intention to intrude into an exclusive federal sphere is neither necessary nor sufficient to scrutinize the applicability of provincial law. It is the fact of intrusion, and not the intention to intrude, which is determinative for division of powers purposes": Husky Oil Operations Ltd v Minister of National Revenue, [1995] 3 SCR 453 at para 45, 96 DLR (4th) 495. See ibid at paras 41-45 for discussion usefully contrasting IJI, which concerns the applicability of valid provincial legislation, with colourability, which concerns the legislation's validity. See also note 51 and accompanying text.

(62) "The abstinence of the Dominion Parliament from legislating to the full limit of its powers could not have the effect of transferring to any provincial legislature the legislative power which had been assigned to the Dominion by the Act of 1867": Union Colliery Company of British Columbia v Bryden, [1899] AC 580 at 588, [1899] UKPC 58 (BAILII) [Union Colliery]. See also Canadian Western Bank, supra note 1 at para 44; CORA, supra note 12 at para 52.

(63) See e.g. Canadian Western Bank, supra note 1 at paras 34, 35, 67; PHS, supra note 10 at para 65; Tsilhqot'in, supra note 10 at paras 131, 148; Carter v Canada (AG), 2015 SCC 5 at paras 49-53. But see Friends of the Oldman River v Canada (Minister of Transport), [1992] 1 SCR 3 at 68, 88 DLR (4th) 1; Alberta Government Telephones v Canada (CRTC), [1989] 2 SCR 225 at 275, 15 DLR (4th) 515.

(64) See supra note 1 at para 35 (listing LSBC, supra note 50; Dominion Stores Ltd v The Queen, [1980] 1 SCR 844, 106 DLR (3d) 581 [Dominion Stores]; and Labatt Breweries of Canada Ltd v Canada (Attorney General), [1980] 1 SCR 914, 110 DLR (3d) 594 [Labatt]).

(65) But see General Motors, supra note 13 (describing Dominion Stores, supra note 64, as having "struck down federal products standards legislation" at 657). In LSBC, the Court concluded as a matter of statutory interpretation that a decision of Law Society benchers, which was taken under valid legal profession legislation, to prohibit lawyers from advertising publicly did not contravene section 32 of the Combines Investigation Act (as it then was), which prohibited conspiracy to restrain or injure competition unduly. See LSBC, supra note 50 at 359. Having done so, the Court saw no need to answer the constitutional question posed about section 32. See LSBC, ibid at 361-62. According to General Motors, the Court in Labatt, supra note 64, struck down altogether federal regulations purporting to establish minimum standards for food products marketed as light beet, because regulation within a province of a single trade or business does not come within federal legislative authority. See General Motors, supra note 13 at 678.

(66) See Elliot, supra note 8 at 469.

(67) [1988] 2 SCR 680, 54 DLR (4th) 679 [Clark].

(68) 2006 SCC 41 [Isen].

(69) Elliot, supra note 8 at 469. In a much earlier case commentary, Elliot lists several earlier decisions that he believes suggest that "[f]ederal legislation is as susceptible to being 'read down' as provincial legislation": Robin Elliot, "Constitutional Law--Division of Powers--Interjurisdictional Immunity, Reading Down and Pith and Substance: Ontario Public Service Employees Union v. Attorney-General for Ontario" (1988) 67:3 Can Bar Rev 523 at 542. These decisions ate Quebec North Shore Paper Co v Canadian Pacific, [1977] 2 SCR 1054, 71 DLR (3d) 111 [Quebec North Shore]; Reference Re Industrial Relations and Disputes Investigation Act, [1955] SCR 529, [1955] 3 DLR 721 [Stevedore Reference]; and LSBC, supra note 50.I don't find these examples as germane to the present issue. In Quebec North Shore, the Court held that section 101 of the Constitution Act, 1867, supra note 29, authorized Parliament to confer on the Federal Court jurisdiction only over matters that were the subject of valid federal laws, not also over other matters coming within federal legislative authority. It had nothing to do with the legislative authority of the provinces. And the Stevedore Reference considered the validity of federal legislation aimed exptessly at the labour relations of undertakings and industries that are within federal legislative authority. In considering the list of undertakings and industries enumerated for greater certainty in the legislation, the Court construed them ejusdem generis in a manner consistent with its understanding of the scope of federal legislative authority. As for LSBC, see supra note 65. Bruce Ryder too finds these examples unconvincing. See Ryder, "Equal Autonomy", supra note 32 at 582-85.

(70) Dominion Stores, supra note 64 at 865-66.

(71) RSC 1970, c R-2, s 342(1), as repealed by Canada Transportation Act, SC 1996, c 10, s 185.

(72) Clark, supra note 67 at 694-95.

(73) Ibid at 708. See generally ibid at 708-09 (for a more complete picture of the Court's argument).

(74) Ibid at 710.

(75) RSC 1985, c S-9, as repealed by Canada ShippingAct, SC 2001, c 2.

(76) Isen, supra note 68 at para 14.

(77) See ibid at paras 23-27.

(78) In the words of the majority of the Court in Canadian Western Bank, IJI's "application to federal laws in order to avoid encroachment on provincial legislative authority has often consisted of 'reading down' the federal enactment or federal power without too much doctrinal discussion": supra note 1 at para 35.

(79) My former student Raeya Jackiw noticed this a little before I did. Kudos.

(80) I overlook for present purposes section 94A of the Constitution Act, 1867, which explicitly gives provincial legislation priority over federal in relation to "old age pensions and supplementary benefits, including survivors' and disability benefits irrespective of age": Constitution Act, 1867, supra note 29, s 94A.

(81) 2001 SCC 67 [Mangat].

(82) Ibid at para 52. See also ibid at paras 53-54. Cf Lafarge, supra note 10 (IJI "should not be used where, as here, the legislative subject matter (waterfront development) presents a double aspect" at para 4).

(83) COPA, supra note 12 at paras 54-59.

(84) See supra note 1 at para 77.

(85) Ibid at para 78.

(86) Supra note 10 at para 49.

(87) Supra note 12 at para 61.

(88) PHS, supra note 10 at para 65. See also Quebec (AG) v Canada (Human Resources and Social Development), 2011 SCC 60 ("[t]his hierarchy [considering paramountcy before, or instead of, IJI] was established as a matter of judicial policy" at para 11).

(89) See Ryan Estate, supra note 10 at para 49; Rogers, supra note 12 at para 63.

(90) See COPA, supra note 12 at para 37.

(91) See supra notes 64-77 and accompanying text.

(92) Canadian Western Bank, supra note 1 at para 78, quoted at greater length in the text accompanying note 85, above.

(93) In PHS, the Court did acknowledge that "new areas of exclusive jurisdiction could in theory be identified in the future", but hastened to add that "courts are reluctant to identify new areas where [IJI] applies": PHS, supra note 10 at para 67.

(94) Supra note 1 at para 78.

(95) As Robin Elliot observed some time ago, the Court's recent preference for paramountcy over IJI "only makes sense if the doctrine of [IJI] is invoked against provincial legislation": Elliot, supra note 8 at 495. This is another indication that the Court has not thought through carefully the implications of invoking IJI to protect provincial authority.

(96) Interprovincial Co-operatives Ltd v The Queen, [1976] 1 SCR 477 at 512, 53 DLR (3d) 321 [Ipco].

(97) Constitution Act, 1867, supra note 29, s 91.

(98) Ibid.

(99) See e.g. Union Colliery, supra note 62 at 585, 587; Reference re Alberta Statutes, [1938] SCR 100 at 115, [1938] 2 DLR 81 [Alberta Statutes], aff'd Alberta Taxation Reference, supra note 51; Canadian Pacific Railway v British Columbia (AG), [1950] 1 DLR 721 at 727-28, [1950] AC 122 (PC); Canada (AG) v The Canadian Pacific Railway Company and Canadian National Railways, [1958] SCR 285 at 290, 301, 12 DLR (2d) 625; Murphy v Canadian Pacific Railway Company, [1958] SCR 626 at 632, 15 DLR (2d) 145; Canada (AG) v Canadian National Transportation Ltd, [1983] 2 SCR 206 at 226-28, 3 DLR (4th) 16. In the words of William Lederman, the opening and closing words of section 91, also known as the declaratory and deeming clauses, "were designed to ensure that the twenty-nine specific categories in the original federal list were to be taken as withdrawn from the historic scope of the provincial property and civil rights clause, and withdrawn also from the new provincial category of things generally of a local and private nature in the province": William Lederman, "University and Diversity in Canadian Federalism: Ideals and Methods of Moderation" (1975) 53:3 Can Bar Rev 597 at 602, quoted in Saywell, supra note 47 at 324, n 38.

(100) Constitution Act, 1867, supra note 29, s 91.26.

(101) Ibid, s 92.12.

(102) Ibid, s 91.3.

(103) Ibid, s 922.

(104) Ibid, s 91.2.

(105) See Lawson v Interior Tree Fruit and Vegetables Committee of Direction, [1931] SCR 357, [1931] 2 DLR 193 ("[t]he scope which might be ascribed to head 2, section 91 (if the natural meaning of the words, divorced from their context, were alone to be considered), has necessarily been limited, in order to preserve from serious curtailment, if not from virtual extinction, the degree of autonomy which, as appears from the scheme of the Act as a whole, the provinces were intended to possess" at 366). See also Canadian Western Bank, supra note 1 at para 43; Reference re Pan-Canadian Securities Regulation, 2018 SCC 48 at para 100 [Pan-Canadian Securities].

(106) See e.g. Reference re Marriage Legislation in Canada, [1912] AC 880 (PC), [1912] UKPC 63 (BAILII) (discussing section 91.26); Lambe, supra note 51 at 585 (discussing section 91.3).

(107) See Reference re Farm Products Marketing Act, [1957] SCR 198 at 209, 7 DLR (2d) 257, quoted with approval in Dominion Stores, supra note 64 (the relevant provincial "power is a subtraction from the scope of the language conferring on the Dominion by head 2 of s. 91 exclusive authority to make laws in relation to the regulation of trade and commerce" at 862); Citizens' Insurance Company v Parsons (1881), 7 AC 96 (PC), [1881] UKPC 49 (BAILII).

(108) Bell 88, supra note 33 at 841.

(109) Natural Parents v Superintendent of Child Welfare, [1976] 2 SCR 751 at 761, 60 DLR (3d) 148 [Natural Parents]. See also ibid ("[i]t cannot be... that because a provincial statute is general in its operation, in the sense that its terms are not expressly restricted to matters within provincial competence, it may embrace matters within exclusive federal competence" at 760).

(110) Elliot, supra note 8 at 493.

(111) COPA, supra note 12 at para 48.

(112) Ibid at para 53. See also ibid at paras 48, 60. In principle, the federal order could disallow inconvenient but valid provincial laws. See Constitution Act, 1867, supra note 29, ss 56, 90. But it could do so, if at all, only within one year of the date the provincial Act received royal assent. And just imagine the hullaballoo that would ensue today if it tried.

(113) In Rogers, the majority held the relevant municipal resolution to be constitutionally invalid, but considered and approved IJI as an alternative ground. Gascon J, concurring in the result, disagreed with the majority on the issue of validity but agreed that IJI rendered the municipal instrument constitutionally inapplicable. See Rogers, supra note 12.

(114) See e.g. Canadian Pioneer Management Ltd v Saskatchewan (Labour Relations Board), [1980] 1 SCR 433, 107 DLR (3d) 1, Laskin CJC ("there is no accretion to provincial legislative authority by the failure or unwillingness of Parliament to legislate to the full limit of its powers under s. 91" at 440).

(115) See e.g. Roberts v Canada, [1989] 1 SCR 322, 57 DLR (4th) 197 [Roberts] (the law of Aboriginal title is federal common law); Ordon Estate, supra note 3 (maritime negligence law is federal common law).

(116) The paramountcy doctrine, as currently understood, would not suffice for this purpose. "Federal paramountcy applies where there is an inconsistency between a valid federal legislative enactment and a valid provincial legislative enactment. The doctrine does not apply to an inconsistency between the common law and a valid legislative enactment": Ryan Estate, supra note 10 at para 66. But see Bisaillon v Keable, [1983] 2 SCR 60 at 108, 2 DLR (4th) 193.

(117) Pan-Canadian Securities, supra note 105 at para 131 [emphasis in original]. See also Comeau, supra note 30 at para 87, quoted in greater length in note 151.

(118) Reasonable people may differ, of course, about whether this is a good thing or a bad thing. My only claim is that preserving Parliament's freedom not to legislate is part of IJI's unique contribution to Canadian division of powers jurisprudence. Considering when, or even whether, Canadian constitutional law should protect Parliament's freedom not to legislate is part of the normative inquiry into IJI's proper place within that law.

(119) See supra notes 15-20 and accompanying text.

(120) Supra note 1 at para 46.

(121) Bell 88 considered and rejected a similar argument almost twenty years before. See Bell 88, supra note 33 at 843-44.

(122) COPA, supra note 12 at paras 62-74.

(123) Ibid at paras 50-60. See also supra notes 111-112 and accompanying text.

(124) See supra note 63 and accompanying text.

(125) See Elliot, supra note 8 at 489-90.

(126) Secsupra note 10 at para 148.

(127) See supra notes 35-50 and accompanying text.

(128) See also Elliot, supra note 8 at 490; Hogg & Godil, supra note 9 at 637.

(129) See supra notes 111-116 and accompanying text.

(130) Sec supra note 16 and accompanying text.

(131) Constitution Act, 1982, supra note 28, s 35 (recognizing and affirming the existing Aboriginal and treaty rights of the Aboriginal peoples of Canada).

(132) See supra note 29 and accompanying text.

(133) Robin Elliot noticed this first. See Elliot, supra note 8 at 486.

(134) See PHS, supra note 10 at para 69.

(135) "[T]here can be no doubt that under that organic instrument [the Constitution Act, 1867], the powers distributed between the Dominion on the one hand and the provinces on the other hand cover the whole area of self-government within the whole area of Canada": Ontario (AG) v Canada (AG) (1912), 3 DLR 509 at 510-11, [1912] AC 571 (PC). See also Reference re Same-Sex Marriage, 2004 SCC 79 at para 34; Comeau, supra note 30 at para 72.

(136) Quebec (AG) v Canada (AG), 2015 SCC 14 at para 44 [Quebec v Canada].

(137) Leaving aside the limits that sections 96 and 121 of the Constitution Act, 1867, the Charter, and section 35 of the Constitution Act, 1982 impose alike on both orders of government. See supra notes 27-30 and accompanying text.

(138) By way of example, there was very little statute law governing eligibility to marry until 2005, when Parliament passed the Civil Marriage Act, SC 2005, c 33--only the Marriage (Prohibited Degrees) Act, SC 1990, c 46, which prescribes which relatives, by blood or adoption, may not marry one another. In the absence of legislation, courts had and used clear common law rules to resolve disputes about marriage eligibility and validity.

(139) In recent years, Parliament has done just that on at least two occasions, enacting legislation to address perceived regulatory gaps on Indian reserves. See First Nations Commercial and Industrial Development Act, SC 2005, c 53 (offering a predictable regulatory framework lor commercial and industrial developments on participating reserves); Family Homes on Reserves and Matrimonial Interests or Rights Act, SC 2013, c 20 (governing Indians' matrimonial property on reserve, where provincial matrimonial property legislation is constitutionally inapplicable; see Derrickson v Derrickson, [1986] 1 SCR 285,26 DLR (4th) 175 [Derrickson]).

(140) Sec e.g. Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 SCR 297, 8 DLR (4th) 1 ("it is not for this Court to consider the desirability of legislation from a social or economic perspective where a constitutional issue is raised" at 334); Comeau, supra note 30 ("[t]he question for a court is squarely constitutional compliance, not policy desirability" at para 83); Pan-Canadian Securities, supra note 105 ("policy considerations and practical effects are irrelevant to the question before this Court" at para 82). See also ibid, paras 68, 70, 130.

(141) Seesupra notes 128-129 and accompanying text.

(142) Canadian Western Bank, supra note 1 at para 24; Tsilhqot'in, supra note 10 at para 149.

(143) Supra note 10 at para 63.

(144) Pan-Canadian Securities, supra note 105 at paras 17-18 [emphasis added]. See also note 168 and accompanying text.

(145) See Peralta v Ontario, [1988] 2 SCR 1045, aff'g (1985), 49 OR (2d) 705 (CA).

(146) See Criminal Code, RSC 1985, c C-46, s 207, as amended; R v Furtney, [1991] 3 SCR 89,129 NR241.

(147) See Pan-Canadian Securities, supra note 105 at paras 117-27.

(148) See Nova Scotia Interdelegation, supra note 40 at 34; Pan-Canadian Securities, supra note 105 at paras 75-76.

(149) See Pan-Canadian Securities, supra note 105 at paras 73, 76, 123-25. See also Sga'nism Sim'augit (Chief Mountain) v Canada (AG), 2013 BCCA 49, leave to appeal to SCC refused (delegation in Nisga'a Treaty of legislative authority to Nisga'a government is constitutionally valid).

(150) See Pan-Canadian Securities, supra note 105 at para 124; R v Dick, [1985] 2 SCR 309, 23 DLR (4th) 33 [Dick]; Derrickson, supra note 139; Coughlin v Ontario Highway Transport Board, [1968] SCR 569, 68 DLR (2d) 384.

(151) "The foundational principle that forms part of the architecture of constitutional texts... is federalism--and that principle does not mandate any specific prescription for how governments within a federation should exercise their constitutional authority": Comeau, supra note 30 at para 87 [emphasis in original].

(152) See e.g. Quebec v Canada, supra note 136.

(153) Elliot, supra note 8 at 489.

(154) See Hogg & Godil, supra note 9 at 636, quoted in the text accompanying note 214, below.

(155) Supra note 1 at para 45.

(156) Ibid.

(157) See to similar effect Elliot, supra note 8 at 496. In this article, Elliot is more convinced than I am at present that IJI operates reciprocally. Cf Ryder, "Equal Autonomy", supra note 32 at 581-87 (IJI does not, at present, operate reciprocally but should).

(158) Constitution Act, 1867, supra note 29, s 91.

(159) See supra notes 63-107 and accompanying text.

(160) Canadian Western Bank, supra note 1 at para 67.

(161) Ibid at para 45.

(162) Ibid.

(163) See supra notes 145-154 and accompanying text.

(164) See supra note 19 and accompanying text.

(165) See supra notes 13, 19 and accompanying text.

(166) "We are not final because we are infallible; we are infallible only because we are final": Brown v Allen, 344 US 443 (1953) at 540, Jackson J (concurring).

(167) Seesupra notes 21-50 and accompanying text.

(168) Reference re Securities Act, 2011 SCC 66 at para 62 [Securities Reference], cited with approval in Quebec v Canada, supra note 136 at para 19, Rogers, supra note 12 at para 39.

(169) Canadian Western Bank, supra note 1 at para 43.

(170) RSC 1985, c I-5.

(171) See ibid, s 4(3), amended SC 2014, c 38, s 4.

(172) See Constitution Act, 1867, supra note 29, s 91.24; Canada (AG) v Canard (1975), [1976] 1 SCR 170, at 193, 202, 52 DLR (3d) 548.

(173) Both Inuit and Metis are "Indians" for purposes of s 91.24. See Reference as to Whether "Indians" in s 91 (24) of the BNA Act Includes Eskimo Inhabitants of the Province of Quebec, [1939] SCR 104, [1939] 2 DLR 417 (Inuit); Daniels v Canada, 2016 SCC 12 (Metis).

(174) General Motors, supra note 13 at 670-72. See ibid at 684. For more on sufficient integration, see note 240.

(175) See Kitkatla, supra note 37 at para 58; Quebec (AG) v Lacombe, 2010 SCC 38 at paras 40-44 [Lacombe].

(176) See supra note 3 and accompanying text. This term appeared originally in Bell 88, supra note 33 at 839.

(177) See supra note 1 at para 85.

(178) Ibid at para 43.

(179) Ibid.

(180) John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971) at 20-21.

(181) Nelson Goodman, Fact, Fiction and Forecast (Indianapolis: Bobbs-Merrill, 1965) ("[t]he process of justification is the delicate one of making mutual adjustments between rules and accepted inferences; and in the agreement achieved lies the only justification needed for either" at 64).

(182) Canadian Western Bank, supra note 1 at para 43.

(183) Ibid.

(184) Cf Elliot, supra note 8 at 494 [footnotes omitted]:
On what basis can it be said that the 'core' or 'basic, minimum and
unassailable content' of Parliament's jurisdiction over federal
elections, maritime law and criminal law, the federal 'activities' to
which [the majority judges in Canadian Western Bank, supra note 1]
refer, is any more or any less difficult to define than the 'core' or
'basic, minimum and unassailable content' oft its jurisdiction over
federally regulated undertakings, 'Indians, and Lands reserved tor the
Indians' and the 'Militia, Military and Naval Service, and Defence'?


(185) Canadian Western Bank, supra note 1 at para 43.

(186) Supra note 10 at para 60.

(187) Supra note 12 at para 30 [emphasis in original].

(188) See ibid at paras 28-29. See also Johannesson v Rural Municipality of West St Paul, [1952] 1 SCR292, [1951] 4 DLR 609 [Johannesson].

(189) PHS, supra note 10 at para 60.

(190) See supra notes 5-6 and accompanying text.

(191) See e.g. COPA, supra note 12 at paras 36-37; PHS, supra note 10 at paras 65, 67; Ryan Estate, supra note 10 at pata 49; Tsilhqot'in, supra note 10 at para 152; Marcotte, supra note 10 at para 63; Rogers, supra note 12 at paras 61-63. See also Canada Post Corporation v Hamilton (City), 2016 ONCA 767 at paras 93-96 [Canada Post]. Thanks to Carol Rogerson for this last reference.

(192) PHS supra note 10 at para 65. Again, the Court did acknowledge that "new areas of exclusive jurisdiction could in theory be identified in the future", but added quickly that "courts are reluctant to identify new areas where [IJI] applies": ibid, at para 67.

(193) The precedent on which the Court in COPA relied most heavily, for example, was Johannesson, supra note 188, a decision from 1952. See COPA, supra note 12 at paras 36-39. And the Court in Rogers invoked IJI on the strength of a Privy Council decision (Corporation of the City of Toronto v Bell Telephone Company of Canada (1904), [1905] AC 52 (PC), 13 CRAC 361) from the days of watertight compartments. See Rogers, supra note 12 at para 6.

(194) See Morris, supra note 38. The oral hearing in the Morris appeal took place on 14 October 2005, roughly six months before the date of the oral hearing in Canadian Western Bank, supra note 1 (11 April 2006). The Court released its decision on Morris on 21 December 2006; Canadian Western Bank came out on 31 May 2007.

(195) Tsilhqot'in, supra note 10 at para 150. The Court did so there, it said, because "Morris... was decided prior to this Court's articulation of the modern approach to [IJI] in Canadian Western Bank and [COPA], and so is of limited precedential value on this subject as a result": ibid. But see supra note 194. In this respect, Morris was, of course, no different from any of the other pre-2007 Supreme Court or Privy Council precedents invoking IJI, including those that the Court relied on in COPA and Rogers. See supra note 193 and accompanying text. Ryan Estate strongly suggests that Ordon Estate, supra note 3, a 1998 decision, would be decided diffetently today. See Ryan Estate, supra note 10 at paras 60-64. For further discussion of the latter issue, see notes 222-231 and accompanying text.

(196) See Delgamuukw v British Columbia, [1997] 3 SCR 1010 at paras 178, 181, 153 DLR (4th) 193.

(197) See Tsilhqot'in, supra note 10 at paras 135-38. Tsilhqot'in has caused considerable confusion about what, if anything, remains at the core of section 91.24 of the Constitution Act, 1867 ("Indians, and Lands reserved for the Indians"). See Kerry Wilkins, "Life Among the Ruins: Section 91(24) After Tsilhqot'in and Grassy Narrows" (2017) 55 Alta L Rev 91 at 112-22. For more general critical discussion of Tsilhqot'ins treatment of division of powers issues, sec Wilkins, ibid; Kent McNeil, "Aboriginal Title and the Provinces After Tsilhqot'in Nation" (2015) 71 SCLR (2d) 67.

(198) See supra note 195. And in Canada Post, the Ontario Court of Appeal displays, without explanation, still greater appetite for selectivity. "It is not sufficient that [IJI] has in the past been applied to the head of power of postal services. There needs to be a closer nexus with the actual subject matter of the By-Law": Canada Post, supra note 191 at para 95.

(199) Supra note 1 at para 67.

(200) See supra notes 169-188 and accompanying text.

(201) See supra notes 187-188 and accompanying text.

(202) Supra note 1 at para 77.

(203) Ibid at para 78, quoted in relevant part in the text accompanying notes 6, 85, above.

(204) See COPA, supra note 12 ("generally" at para 26, "in general" at para 36); Marcotte, supra note 10 ("in general" at para 63); Rogers, supra note 12 ("generally" at para 61). But see PHS, supra note 10 ("before applying the doctrine of [IJI] in a new area, courts should ask whether the constitutional issue can be resolved on some other basis" at para 65).

(205) See the text accompanying note 5, above.

(206) Supra note 1 at para 77, cited with approval in Ryan Estate, supra note 10 at para 50, Rogers, supra note 12 at para 61.

(207) Supra note 12 at para 35. See also ibid citing Canadian Western Bank, supra note 1 ("[t]he core of a federal power is the authority that is absolutely necessary to enable Parliament 'to achieve the purpose for which exclusive legislative jurisdiction was conferred'" at para 35 [citations omitted]).

(208) Supra note 12 at para 36.

(209) Rogers, supra note 12 at para 61.

(210) Moloney, supra note 44 at pata 14.

(211) Canadian Western Bank, supra note 1 at para 22.

(212) 114957 Canada Ltee (Spraytech, Societe d'arrosage) v Hudson (Town), 2001 SCC 40 at para 3 [Spraytech].

(213) See e.g. Ipco, supra note 96, Pigeon J ("[s]uch authority is under every head expressly or impliedly restricted to the provincial territory" at 512-13).

(214) Hogg & Godil, supra note 9 at 636.

(215) See Simeon, supra note 43 at 135-41. It is noteworthy that the only two occasions since Canadian Western Bank on which the Supreme Court invoked IJI--COPA and Rogers--each limited the reach of legislation from Quebec. See COPA, supra note 12; Rogers, supra note 12.

(216) Simeon, supra note 43 at 141. See generally ibid at 135-55.

(217) By way of example, we know that the framers' decision to place "Indians, and Lands reserved for the Indians" within federal legislative authority was deliberate; immediately before Confederation, that power resided with the federating provinces. See PG McHugh, Aboriginal Societies and the Common Law: A History of Sovereignty, Status, and Self-Determination (Oxford: Oxford University Press, 2004) at 182-83. The better explanation of the reasons for that decision is that "Indians" were thought to need the protection of the more distant level of government from the conflicting interests of the local settlers. For further discussion and source material, see Wilkins, supra note 197 at 95-98.

(218) See note 226.

(219) See Roberts, supra note 115.

(220) This approach to IJI calls to mind certain features of the Supreme Court's discussion in General Motors of federal legislative authority over general regulation of trade. See General Motors, supra note 13. The validity of a federal scheme to regulate trade throughout Canada, the Court said there, requires, among other things, that the legislation be "of a nature that the provinces jointly or severally would be constitutionally incapable of enacting" and that "the failure to include one or more provinces or localities in a legislative scheme would jeopardize the successful operation of the scheme in other parts of the country": ibid at 662. The two discussions agree that, in respect of some matters, the risk of provincial or local variation in approach is sufficient, and sufficiently unwelcome, to be dispositive. There are, however, some important differences between the two discussions, most of which stem from the fact that IJI is about ascertaining possible limits on the reach of valid provincial legislation, not about testing any federal legislation for validity. IJI operates even in the absence of relevant federal legislation because exclusive federal legislative authority persists irrespective of whether Parliament chooses to use it. See supra note 62 and accompanying text. A test for IJI, therefore, must not presuppose the existence or contemplation of a relevant federal legislative scheme; it must seek instead to identify the kinds of matters that require a measure of immunity from valid, potentially disparate provincial legislative schemes. And the problem IJI seeks to solve is not provincial incapacity validly to achieve sufficient uniformity but the Constitution's incapacity otherwise to ensure, in matters of sufficient national importance, that any countrywide consensus reached through multiple independent, valid provincial initiatives would endure.

(221) Supra note 1 at para 48, quoted at greater length in the text accompanying note 7, above.

(222) Supra note 3.

(223) Supra note 10.

(224) Sec Ordon Estate, supra note 3 at para 85; Ryan Estate, supra note 10 at para 59 (citing this passage with approval).

(225) See Constitution Act, 1867, supra note 29, s 91.10; Ordon Estate, supra note 3 at para 84; Ryan Estate, supra note 10 at para 59.

(226) See Ordon Estate, supra note 3 at para 88:
The attribution to Parliament of exclusive legislative jurisdiction
over navigation and shipping stems in large part, in our view, from the
national and international dimensions of maritime law, and the
corresponding requirement for uniformity in maritime law principles. If
matters of maritime law were regulated by the various provincial
legislatures, this would drastically confuse the day-to-day reality of
navigation and shipping in Canadian waters, and would make it
impossible for Canada as a country to abide by its international treaty
obligations relating to maritime matters[.]


See also Ryan Estate, supra note 10 at para 61.

(227) Ordon Estate, supra note 3 at paras 85, 87.

(228) Ibidzt para 109.

(229) Ryan Estate, supra note 10 at para 62.

(230) Ibid at para 64.

(231) Ibid at para 62. This conclusion suggests a standard of uniformity for maritime negligence law substantially less robust than that articulated in Ordon Estate, supra note 3. See supra note 226. One might have thought, for example, that the provincial legislation at issue in Ryan Estate did compromise "the scope of available damages, and the availability of a regime of apportionment of liability according to fault": Ordon Estate, supra note 3 at para 85, quoted in the text accompanying note 224, below. The Court in Ryan Estate thought it sufficient that "parties in the position of the Ryan Estates still receive compensation for the accident in question (albeit through a different mechanism and from a different source)": Ryan Estate, supra note 10 at para 62. This is true, but both the measure of compensation payable and the basis for compensation (no fault, not "apportionment of liability according to fault") are quite different: Ryan Estate, supra note 10 at para 52; Ordon Estate, supra note 3 at para 85.

(232) Sec COPA, supra note 12 at paras 66-74.

(233) Marcotte, supra note 10 at para 69.

(234) See Rogers, supra note 12 at para 71.TheCourtin Rogers, having decided the case on other grounds, declined to consider the paramountcy issue. See ibid at para 74.

(235) Marcotte, supra note 10 at para 69.

(236) See supra notes 108-110 and accompanying text.

(237) See supra notes 210-220 and accompanying text.

(238) Contra Canadian Western Bank, supra note 1 at para 48, quoted in the text accompanying note 7, above.

(239) See e.g. Four B Manufacturing v United Garment Workers (1979), [1980] 1 SCR 1031 at 1047-48,102 DLR (3d) 385.

(240) See COPA, supra note 12 ("'[i]mpairment'... suggests an impact that not only affects the core federal power, but does so in a way that seriously or significantly trammels the federal power.... [A]pplication of the doctrine of [IJI] requires a significant or serious intrusion on the exercise of the federal power" at para 45). There is one circumstance in validity inquiry in which courts have taken account of the degree to which a provision in provincial legislation encroaches on federal authority, or vice versa. Under the so-called ancillary powers doctrine, a provision in provincial (or federal) legislation that, considered in isolation, appears to reach beyond the legislative authority of the enacting order of government can nonetheless be accepted as valid it it is sufficiently integrated into an otherwise valid provincial (or federal) scheme. The test for sufficient integration has been said to vary with the extent to which the impugned provision encroaches on the legislative authority of the other order of government; the more severe the perceived encroachment, the more stringent the requirements for sufficient integration. See e.g. General Motors, supra note 13 at 666-72; Kitkatla, supra note 37 at para 58; Lacombe, supra note 175 at paras 40-44. Unfortunately, the Court has yet to tell us how to gauge the severity of any such intrusion; as it acknowledged in Lacombe, "one's view of the seriousness of an intrusion may vary depending on whether one is intruding or being intruded upon": Lacombe, supra note 175 at para 43. In practice, the Court has almost always placed the intrusion, without explanation, at the lower end of the severity scale. See Eugenie Brouillet, "Canadian Federalism and the Principle of Subsidiarity: Should We Open Pandora's Box?" (2011) 54 SCLR (2d) 601 at 622-24. Robin Elliot, writing in 2008, recommended "requiring a court to end its analysis and strike down the impugned part of the statute if it found that that part, viewed in isolation, encroached on a core aspect of a head of power assigned to the other order of government": Elliot, supra note 8 at 494. See also Reference re Assisted Human Reproduction Act, 2010 SCC 61 at para 188, LeBel & Deschamps JJ (for four of nine-judges) [Reproduction Reference].

(241) Sec Bell 88, supra note 33.

(242) See Ordon Estate, supra note 3. But see now Marine Liability Act, SC 2001, c 6, ss 4, 6 (which include siblings of the injured or deceased among those who may sue in such situations).

(243) Sometimes, however, such outcomes give rise to appropriately corrective federal legislative action. See e.g. supra notes 139, 242. See also Ben Depoorter, "The Upside of Losing" (2013) 113 Columbia L Rev 817.

(244) See Securities Reference, supra note 168.

(245) See Reproduction Reference, supra note 240.

(246) Sec Reference re Senate Reform, 2014 SCC 32.

(247) Reference re Manitoba Language Rights, [1985] 1 SCR 721, 19 DLR (4th) 1.
COPYRIGHT 2019 University of British Columbia Law Review Society (Canada)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wilkins, Kerry
Publication:University of British Columbia Law Review
Geographic Code:1U3IN
Date:Jun 1, 2019
Words:23315
Previous Article:CHILD PLACEMENT AND THE LEGAL CLAIMS OF FOSTER CAREGIVERS.
Next Article:CONSTITUTIONAL OR POLITICAL CRISIS ? PROSECUTORIAL INDEPENDENCETHE PUBLIC INTEREST AND GENDER IN THE SNC-LAVALIN AFFAIR.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters