I. Behind the written word are the Constitution’s "vital unstated assumptions". If it is impossible to conceive of the constitutional structure without them, as the Supreme Court of Canada said in the Secession Reference, their role in interpretation is dubious just the same. Though its meaning may not be self-evident, the text identifies the principles that are entrenched as the supreme law of our land. Unstated assumptions which might be considered vital cannot claim the pedigree of text or the supreme status it confers.

In deference to the text, the Supreme Court has acknowledged that "there are compelling reasons to insist upon the primacy of our written constitution". At times, then, it is suggested that courts engaging in review do no more than enforce the commands of the text. Insofar as the object of review is to "divine" the purpose or intent of constitutional documents, the exercise is "akin to statutory interpretation". From that perspective a textual link is imperative, for courts asserting power against democratic institutions under an "authoritative constitutional instrument" must respect the "democratic foundation of judicial review" - that "super-legislative source" that grounds the authority to nullify democratically enacted laws in the Constitution’s written word. When the judiciary strays outside the four corners of the document, its legitimacy to exercise that power is placed at risk.

Still, the parsing of words cannot be avoided, and in that regard strict textualism is at once too rigid and too inconclusive to serve as a constitutional methodology. In the Court’s words, "[a]lthough these texts have a primary place in determining constitutional rules, they are not exhaustive". To put it candidly, judge-made rules and principles are inescapable. Under the division of powers, for example, the text does not limit the federal government’s power to legislate for the peace, order and good government to emergencies, or confine its actions under the general power by reference to a multiple part national dimensions test. The desirability of such constraints, not to mention how each might apply in a variety of settings, cannot be drawn from the opening words of s.91 in the Constitution Act, 1867. By the same token, s.1's undefined concept of reasonable limits and the structured, judge-made test that emerged in R. v. Oakes could hardly be more foreign to one another. Even so, Oakes can be faulted as mechanistic and formalistic, or applauded for its insistence that limits on constitutional rights be subject to a strict standard of justification. In either case it is clear that, as judicial longhand for the Charter’s requirement of reasonable limits, the Oakes criteria are extra-textual. Examples which can be multiplied many times over, under the division of powers and the Charter, would confirm that the judiciary makes extraordinary choices in interpreting the supreme law of the land.

If the written word is not conclusive, nor are the courts free to disregard its conscious choices. In other words, doubts about the legitimacy of review do not evaporate with the failure of strict textualism, and if the text is not dogma, nor is an absolute prerogative of judicial interpretation. The challenge is to mark out a position that falls somewhere between the rigidity of textualism and the indeterminacy of the Constitution’s vital unstated assumptions.

This collection of essays honours Senator Gérald-A. Beaudoin, whose accomplishments over an exceptional professional life have enriched the legal and political institutions of this country. It is fitting, in light of his contributions both as a lawyer, scholar and dean of law, and as a member of the Senate of Canada, to address the relationship between the judicial and parliamentary branches of our constitutional order. By asking whether courts are entitled to rely on unwritten principles to place limits on legislative authority, this paper considers one aspect of that dynamic. An initial section frames the issue in blunt terms by asking whether the Supreme Court of Canada has engaged in lochnering, that reviled doctrine of American constitutionalism. Accepting that Lochner v. New York was the product of certain historical and cultural dynamics, the example nonetheless places the relationship between the written word and the legitimacy of review in sharp focus.

The two sections which follow the discussion of lochnering examine the Supreme Court of Canada’s reliance on unwritten principles to interpret the Constitution in the Provincial Judges Reference and the Secession Reference. In the first, the Court discounted the express provisions of the text in concluding that the preamble, as the "grand entrance hall to the castle of the Constitution", is the "true source" of a substantive commitment to judicial independence. Granted, it is part of the Constitution Act, 1867; still, the preamble had not previously been considered a source, much less the true source, of significant constraints on government authority. Not long thereafter, the Supreme Court provided answers to three questions posed by the federal government in the Secession Reference. Though ostensibly decided under the Constitution, the Court’s response to the hypothetical of Quebec independence relied on a series of unwritten principles in proposing that the rest of Canada has a constitutional duty to negotiate the terms of secession. It may have been unfair to ask the Court for a pronouncement on the politically charged question of secession, which has no textual presence in the Constitution. Even so, little in the Court’s response can claim any connection, either to the text or to established principles of constitutional law.

After discussing the Supreme Court’s two key decisions in some detail, the paper explores the irrelevance of the text in this jurisprudence, and explains why that is a troubling development. Unfortunately, it is not possible, under the constraints of this paper, to offer a more developed conception of the relationship between the text and all manner of unwritten sources of principle. Though some may be more valid as a matter of interpretation than others, this paper does not attempt to draw that distinction. Its limited objective is to demonstrate that an "unwritten principles" jurisprudence, which completely

ignores the text in interpretation, is not necessary to prevent the Constitution from becoming a strait jacket.

From the beginning, Canadian courts interpreting the Charter have been aware that lochnering refers to a style of interpretation which prevailed during a dark period in American constitutionalism. The term takes its name and stigma from Lochner v. New York and the jurisprudence it spawned. Decided in 1905, Lochner constitutionalized liberty of contract in employer-employee relations to defeat progressive legislation aimed at a variety of exploitative practices that accompanied a period of rapid industrialization and economic growth. In doing so, the U.S. Supreme Court acted on its own unstated assumptions about the primacy of economic liberty, at least against the preferences of democratically enacted legislation. This style of interpretation remained controversial until 1937, when the saga of laissez-faire constitutionalism ended in the catharsis prompted by President Roosevelt’s court-packing plan. As a matter of self-preservation the U.S. Supreme Court hastily retreated from an interpretation of the due process clause that protected economic liberties. When its composition changed, members of the post-crisis court did not hesitate to declare the institution’s scorn for lochnering.

From an American perspective, the lochner jurisprudence undercut the legitimacy of review for several reasons. One was that this case law effectively granted laissez-faire values a constitutional immunity from legislative regulation. Moreover, not only did the U.S. Supreme Court give the due process clause a substantive interpretation, by protecting entitlements the text did not mention, it inevitably made certain values a favourite of the Constitution. In such circumstances, it was difficult to avoid the conclusion that liberty of contract and related interests had been the beneficiaries of subjective judicial preferences. More generally, lochnering compromised the legitimacy of review because the U.S. Supreme Court refused to accept limits on its authority to interpret the Constitution. That is why it was an exercise in exasperation for Stone J. to declare, not long before the court crisis, that "[t]he only check upon our own exercise of power is our own sense of self-restraint". One such limit was the text and, in the wake of the Lochner crisis, constitutional interpretation maintained a careful connection with the text for many years thereafter.

In due course, the U.S. Supreme Court once again succumbed to the temptation to constitutionalize extra-textual values. Following signal decisions in Griswold v. Connecticut and Eisenstadt v. Baird, the Court held in Roe v. Wade that the fourteenth amendment protects a woman’s right to seek an abortion. On its face Roe v. Wade bore an unwelcome resemblance to Lochner v. New York: in neither case was there a link between the text of the Constitution and the substantive entitlement enforced by the Court. That connection mattered in Roe v. Wade, because abortion was and still remains controversial in the United States. By removing that issue from democratic control, the U.S. Supreme Court’s decision to treat unwritten values as though they are protected by the text undermined the legitimacy of review. At the least, Roe demonstrated that lochnering’s flaw was not only that the Court of the early 20^th century had chosen the wrong values. Under each of the two lines of jurisprudence, the Court exercised a power of review that exceeded its authority to interpret the text of the Constitution.

In Canada, the American discourse on the legitimacy of judicial review is considered history, time and culture bound. Without acknowledging it by name, Lamer J. said as much about Lochner and U.S. concerns about the undemocratic nature of judicial review in the B.C. Motor Vehicle Reference. There, in considering whether s.7 of the Charter should be given a substantive interpretation, he was quick to declare that Canada would do its Constitution "a disservice" to allow the American debate about the legitimacy of review to govern its interpretation. Though he could have done so, Mr. Justice Lamer did not limit his remarks to the interpretation of s.7, but pronounced, more generally, that "[a]djudication under the Charter must be approached free of any lingering doubts as to its legitimacy". In his view, the courts could not be faulted for enforcing the Charter when the elected representatives extended the scope of constitutional adjudication and "entrusted the courts with this new and onerous responsibility".

This paper does not consider whether s.7 or the due process clause should be given a substantive interpretation. It focuses, instead, on a more troubling development in constitutional interpretation: the emergence of an "unwritten principles" jurisprudence. Concerns about the legitimacy of adding substantive content to express constitutional provisions, like s. 7 of the Charter or the fourteenth amendment’s due process clause, apply to the unwritten principles jurisprudence in this way. Accepting that there are differences to take into account, the Lochner and Roe jurisprudence demonstrates that the judiciary’s authority to interpret the Constitution is linked to the text. Yet by focusing on differences at the expense of similarities, the Motor Vehicle Reference obscured that link. It is true that the Charter was created by democratic institutions, and that Canada and the United States are not the same. Still, Mr. Justice Lamer’s dismissal of American experience was short-sighted, for it discounted the lessons the Supreme Court of Canada could have learned about the relationship between constitutional texts and the legitimacy of review. As the discussion of the Provincial Judges Reference and the Secession Reference shows, the Supreme Court of Canada has not yet realized that the legitimacy of review is not free of lingering doubts, or that the text operates as a limit on constitutional interpretation.

The Provincial Judges Reference comprises four cases which were consolidated on the question whether the government’s power to set or alter the provincial judiciary’s salary violates s.11(d) of the Charter. In answer, a majority led by then Chief Justice Lamer held that a general principle of judicial independence is protected by the Constitution, and that the relationship between the courts and the other branches of government, must, as a matter of constitutional imperative, be "depoliticized". Thus he claimed that "the possibility of, or the appearance of, political interference through economic manipulation" could only be thwarted by a special process under the Constitution. To avert that risk, the Court held that the provinces must establish "independent judicial commissions" to protect the financial security of provincial court judges.

The Provincial Judges Reference is an easy target for criticism. The Court decided a question that was not briefed or argued by the parties, the impropriety of which greatly irritated LaForest J., who commented on it repeatedly in his dissenting reasons. Moreover, having concluded elsewhere that the Constitution does not protect economic entitlements, the Court managed to entrench the financial security of provincial judges. In order to do so it discovered a substantive commitment to judicial independence in the preamble to the Constitution Act, 1867. After creating the entitlement and discussing the required independent judicial commissions in some detail, the Court denied that it had laid a "particular institutional framework" down in "constitutional stone". Instead, the Chief Justice maintained that the Court’s proposal for an "institutional sieve" between the judiciary and their governments merely offered a template or set of guidelines to assist provincial legislatures in designing institutional arrangements for the protection of judicial independence.

In this discussion the question of judicial independence is less important than the Court’s conception of "unwritten principles". To begin, it should be noted that the Constitution nowhere identifies financial security as an entitlement the judiciary can enforce against either level of government. As for independence, the principle is mentioned but once, in s.11(d)’s guarantee that those accused of offences are entitled to be tried by an "independent and impartial tribunal". Patently, the purpose of the provision is to protect those who are subject to the authority of judges, and not to confer an entitlement on the judges themselves. Elsewhere, the status of judges is addressed in ss.96 to100 of the Constitution Act, 1867. Under the division of powers, s.100 establishes that the federal government is responsible for the salaries of the judges it appoints. Otherwise, given s.92(14)’s deference to the provinces in matters pertaining to the administration of justice, provincial courts are neither established by nor mentioned in the Constitution.

Absent recognition of judicial independence in either text, the consolidation of cases that arrived at the Supreme Court focused on s.11(d) of the Charter, which was the only section that could remotely support a claim to entrench the financial security of provincial court judges. Assuming that a guarantee for the accused can benefit judges, the provision is specific to offences and therefore created a gap in constitutional status for judicial independence in civil proceedings. Meanwhile, no assistance could be drawn from ss.96 to 100, which apply to federally appointed judges. After admitting s.11(d)’s limits and grumbling that the Court should not be a "prisoner of the case presented", Chief Justice Lamer defied the record and raised the "larger question" whether the general principle of judicial independence has a "constitutional home". His conclusion, that judicial independence is "at root an unwritten constitutional principle ... exterior to the particular provisions of the Constitution", prompted Mr. Justice La Forest to write a dissenting opinion that addressed the relationship between the constitutional text and the legitimacy of review.

Although the failure to entrench judicial independence and its corollary of financial security could only be remedied by elevating an unwritten principle to constitutional status, Lamer C.J.C. sought a link to the text in the preamble. First, though, he agreed that "[t]here are many important reasons for the preference of a written constitution over an unwritten one", and acknowledged that, "strictly speaking", the preamble is not a source of positive law. Still, after making those observations, he declared that the opening words of the Constitution have a "special legal effect", which is not limited in use as an aid to interpretation, but "goes even further". In his view, the preamble recognizes and affirms unwritten organizing principles which invite the courts "to fill out the gaps in the express terms of the constitutional scheme".

Once the Chief Justice established that link, it followed that the Court could read a substantive commitment to judicial independence into the preamble. As he explained, there are "serious limitations to the view that the express provisions of the Constitution comprise an exhaustive and definitive code for the protection of judicial independence". In his view it was impossible that provisions which, "by their terms, do not appear to address this objective" might be determinative. Given the "central place that courts hold within the Canadian system of government", he did not consider it feasible that the Constitution might leave judges in a position of financial insecurity.

In this way, the Provincial Judges Reference created a hierarchy, in which unwritten principles became paramount over the express provisions of the Constitution. That hierarchy is inherent in the Chief Justice’s declaration that the preamble is the true source of a foundational principle of judicial independence, as well as in comments suggesting that the Constitution’s substantive provisions are subordinate. Thus, as "expressions of the unwritten principle of judicial independence which is recognized and affirmed by the preamble", neither s.11(d) nor ss. 96 to 100 defined the scope, or lack thereof, of judicial independence under the Constitution. According to Chief Justice Lamer, far from indicating that judicial independence is only "constitutionally enshrined" under the terms of s.11(d), the guarantee is "proof of the existence of a general principle of judicial independence that applies to all courts no matter what kind of cases they hear". The assertion that the specific provisions of the text "merely elaborate" the Constitution’s organizing principles or "underlying logic" can only be viewed as a radical departure from conventional principles of constitutional interpretation.

The Chief Justice also employed a separation of powers rationale to bolster his assertions that financial security is an entitlement and that independent judicial commissions are constitutionally required. Though the judiciary is not entrenched, that did not deter him from declaring the separation of powers to be a "fundamental principle of the Canadian Constitution". In his view, the "political institutions are only one part of the basic structure of the Canadian Constitution", and the courts are equally definitional to the Canadian understanding of constitutionalism. Once again, without authority from the text, he portrayed courts as the co-equal of government. While admitting that the depoliticization of the relationship between the judiciary and the political branches is "largely" governed by convention, rather than by any textual provision or concrete norm, the Chief Justice insisted that the courts must interpret the provisions of the Constitution "in such a manner as to protect this principle". Not only was judicial independence read into the preamble as a result, the Court incorporated an implied guarantee for the independence of provincial courts into s.92(14).

Unwritten principles which are not constrained by the text leave the Court at liberty to define their content. In this instance, the preamble’s commitment to judicial independence enabled the Court to propose extensive guidelines for "independent, effective and objective" judicial salary commissions. It was novel for the Court to find a substantive commitment to judicial independence in the preamble in the first place; the commission mechanism which followed was entirely outside the realm of prior institutional obligations mandated by the Constitution. As if to emphasize the extent to which the mechanism was a creation of the Court, the Chief Justice detailed that the government must respond to the contents of the commission’s report, must do so within a specified period, and must provide a response to the commission’s recommendations. And if those recommendations are not accepted, the government’s decision is subject to judicial review. According to the Chief Justice, this complex sequence of obligations flowed from the preamble and its intention to create a Constitution "Similar in Principle to that of the United Kingdom".

Mr. Justice La Forest’s dissent found fault with the majority opinion on a number of points, including its historical analysis. Significantly, he considered it inappropriate for the majority to base so much of its opinion on that "larger question" whether independence places constraints on the provinces in their dealings with the judiciary. Far from being a "side-wind" which could be left unanswered, the Chief Justice’s remarks were a "lengthy and sustained exegesis" on the issue. In the circumstances, La Forest J. was troubled that parties who had based their submissions on s.11(d) of the Charter had neither briefed nor argued the point. From his perspective, that exegesis was not only a breach of fairness to the litigants but a risky foray that denied the Court the benefit of submissions on an important point of constitutional interpretation.

Mr. Justice La Forest was likewise disturbed by the Chief Justice’s comments about the "proper" relationship between the political and judicial branches, an issue "on which judges can hardly be seen to be indifferent, especially as it concerns their own remuneration". It could appear, he cautioned, that in ignoring the record and entrenching financial security for the provincial judiciary, the Court had set its own agenda and not decided the case impartially. To emphasize the point, La Forest J. provided a reminder that the legitimacy of judicial power must be found in its limits. In his view, by not accepting the limits of the case presented, the Court undermined its own power.

Second, Mr. Justice La Forest rejected the majority’s account of the relationship between the Constitution and its unwritten principles. In particular, he disagreed with the proposition that the court is entitled to fill in the gaps whenever it perceives that the text is not exhaustive. Given that the Constitution named the courts that were given constitutional protection, La Forest J. was not prepared to agree that the failure to address judicial independence in express terms should be regarded as a gap. In his view, the text defined and illustrated institutional relationships, including the distinction between superior and provincial court judges; the difference between federal and provincial jurisdiction under s.96 and s.92(14); and the status of judicial independence in a constitutional order defined by parliamentary supremacy.

He also challenged the Chief Justice’s suggestion that the Constitution’s terms merely elaborate the preamble’s underlying, organizing principles. Given the text’s limited recognition of judicial independence, "it seems strained", La Forest J. stated, "to extend the ambit of this protection by reference to a general preambular statement". He concluded that adding substantive content to the preamble was so unprincipled it would subvert the "democratic foundation of judicial review", which, he added, is "politically legitimate only insofar as it involves the interpretation of an authoritative constitutional instrument". Mr. Justice La Forest indicated quite strongly that the legitimacy of constitutional interpretation "stems from its grounding in an expression of democratic will, not from a dubious theory of an implicit constitutional structure". The express provisions of the Constitution are not elaborations of unwritten assumptions; they are the Constitution. Not only had the majority demeaned the text, in doing so it forgot that the Court cannot impose its views at large; the exercise of review must be related to the text. For that reason, the preamble could not support the obligations a majority of the Court imposed on provincial legislatures.

The Provincial Judges Reference came to the Supreme Court of Canada in the midst of heated disputes between governments and their courts in several provinces. Attentive to that context, the Chief Justice held that independence raised a "live legal issue", in circumstances of "serious strain" on the question of salary, with implications for litigation, which he acknowledged sympathetically, is a "very serious business". Yet in purporting to depoliticize the question of remuneration, the Court politicized its interpretation of the Constitution. As La Forest J. pointed out, constitutionalizing the financial security of judges revealed the Court as a partisan in the cause. He did not consider it plausible that a reasonable person would fear that the fairness of judges depended on commissions to set their salaries. In the end, the Provincial Judges Reference can only be defended on the view that the result is all that mattered.

It is telling that Mr. Justice La Forest was the only member of the panel to harbour lingering doubts about the legitimacy of review. A majority of the Court proceeded as though the text imposed no limits on the Court’s authority to interpret the Constitution.

The Secession Reference addressed the legality, under domestic and international law, of an hypothetical attempt by the province of Quebec to withdraw from Confederation following a majority referendum vote favouring independence. On reading the Court’s answer to the first of three questions posed, which considered the status of Quebec secession under the Constitution, it is difficult to identify familiar principles and sources of authority. That is because the Secession Reference provided a political response in which the law played little or no part.

The federal government’s decision to refer a package of three questions on the secession of Quebec was a kind of pre-emptive strike that placed the Supreme Court of Canada in an untenable position. Those questions were carefully framed to minimize the risk that the federal government might receive unwelcome answers. 3. In the event of a conflict between domestic and international law on the right of the National Assembly, legislature of government of Quebec to effect the secession of Quebec from Canada unilaterally, which would take precedence in Canada? Secession Reference, supra note 1 at 228. To gain ground against the secessionist cause, which very nearly succeeded in the provincial referendum of 1995 was, after all, the point of the exercise. Presumably, though, the Court could not have been pleased about being pressed into service for the federal government on a question of such grave implications for the country. Yet the reference was set up to yield easy answers, and the Court was obligated to respond.

Whether in part to demonstrate institutional independence or not, the Court’s unanimous and anonymous opinion found a way to deny the federal government the absolute victory it sought on the questions of law posed. With no choice but to agree that any unilateral declaration of independence would be illegal under the 1982 amending formula, the Court qualified its answer to the federal government’s first question by adding that a clear, affirmative answer to a clear referendum question would trigger a duty, on the parties to Confederation, to negotiate the independence of Quebec. That obligation is captured, simply enough, in the statement that, "[u]nder the Constitution, secession requires that an amendment be negotiated".

Not surprisingly, the so-called duty to negotiate attracted considerable attention. In raw political terms, this judge-made innovation enabled separatists to claim that the rest of Canada would be obliged to come to terms with a majority vote in Quebec to leave Confederation. Yet any such duty is a novelty that appeared in the Secession Reference without prior lineage. The Constitution is predictably silent on the question of withdrawing from union, and there was no basis on which the preamble to the Constitution Act, 1867 could embrace a substantive obligation to comply with the demand that secession be negotiated. Unlike the principle of judicial independence, which in the Provincial Judges Reference leaned on s.11(d) of the Charter and the jurisprudence under s.100 of the Constitution Act, 1867, this duty could find no home in the text. Nor could it be characterized as a constitutional convention or principle of long-standing practice. In the circumstances, the Court arrived at the duty through the logic that secession could not be achieved, legally, without an amendment to the Constitution, and the process of amendment "perforce requires negotiation". That is how the judges transformed what was otherwise little more than an observation into a duty under the Constitution.

If the federal government prevailed on the formal question of law, the Court’s energy focussed on developing the four unwritten principles that provided the foundation for the duty. To confer a semblance of legitimacy on that obligation, the Court maintained that it derives from four fundamental, organizing principles of the Constitution, which not only establish the duty’s existence but govern the conduct of any party on whom it falls. As to their genesis, the most the Court could say was that "the evolution of our constitutional arrangements has been characterized by" adherence to these principles. By way of rationalization, the Court explained that unwritten principles can create constitutional obligations which are abstract and general, or specific and precise. In either case, such principles generate "substantive legal obligations" which place "substantive limitations upon government action", and are invested with "powerful normative force" that is binding on courts and governments. Although "not explicitly made part of the Constitution by any written provision", the Court continued, "it would be impossible to conceive of our constitutional structure" without such principles. In their absence, the judges intimated, the text would be rigid and unbending, a veritable straitjacket of words.

Thus a duty to negotiate was culled from four constitutive principles the Court discussed at some length: federalism, democracy, constitutionalism and the rule of law, and the protection of minorities. In singling these principles out, the Court emphasized that "the Constitution is more than a written text", and must embrace "the entire global system of rules and principles which govern the exercise of constitutional authority". In such circumstances, the opinion warned, "a superficial reading of selected provisions of the written constitutional enactment, without more, may be misleading". The statement is striking, for it treats the text as irrelevant; that much is evident in its references to a "superficial reading" of "selected provisions", and the inference that the "written" enactment, far from being supreme, is actually "misleading". With those words, the nine judges effectively claimed the authority to embellish and alter the meaning of a document it deemed unreliable.

Accordingly, the opinion continued, it was necessary to make a "more profound investigation of the underlying principles that animate the whole of our Constitution". That investigation produced the four principles above, each of which identifies a familiar theme in Canadian constitutionalism. Yet it is a separate question whether such truisms are a legitimate source of obligations independent of and perhaps even in conflict with the text. Though each of the principles is built into the institutional and substantive framework of the Constitution, the Court suggested that any of them might support a host of as yet undiscovered obligations. The duty to negotiate is a case in point.

The Secession Reference’s unorthodox approach to constitutional interpretation has largely escaped criticism. This, despite agreement among commentators that there is no basis in the constitutional text or in principle for the declaration that the Constitution requires the parties to Confederation to negotiate secession with Quebec. Despite that conclusion, most have been prepared to overlook that vital flaw in the Court’s decision. To some, reference questions which attempted to use the Court in an heavy handed maneuver against Quebec invited a tempering response. As government lawyer Mary Dawson explained, "it is wonderful that Canada is a country where such highly charged issues as the potential secession of one of its provinces can be taken to its highest court". She applauded the Supreme Court of Canada for providing balanced and thoughtful advice in a high-quality opinion, and concluded that even if it the concept was created out of thin air, it would be difficult to quarrel with the legitimacy of the duty to negotiate. Others, like Peter Hogg, suggested that the Court had stated the obvious: if it was inevitable that the federal government would negotiate with Quebec, the Court’s decision simply converted a political reality into a legal rule.

The other view is that the integrity of constitutional interpretation should not be so transparently co-opted by "political reality". The question of Quebec secession was hypothetical at the time of the Reference, and the Court was not confronted by circumstances of necessity requiring an extraordinary response to political reality, as happened in Re Manitoba Language Rights. Accordingly, Daniel Turp and Gibran van Ert have expressed their regret that the Court "built this most significant constitutional development on such a shaky foundation". The Secession Reference earned their disapproval because the Court did not restrict itself to the law but gave extralegal responses; in their view, "it is not the Court’s place to give answers of this type", which detract from sound legal reasoning.

The other problem is that the Secession Reference duty now applies to the entire process of amendment under the Constitution. As the Court itself declared, "[t]he Constitution Act, 1982 gives expression to [the principle of democracy] by conferring a right to initiate constitutional change on each participant in Confederation"; the existence of this right, in turn, "imposes a corresponding duty on the participants in Confederation to engage in constitutional discussions in order to acknowledge and address democratic expressions of a desire for change". Though the nature of the obligation remains sketchy, that problematic has not deterred would-be litigators who have invoked the duty in litigation since the Secession Reference. That it cannot be given determinate content or terms of enforcement is some indication that such matters lie outside the purview of constitutional interpretation and that the Court has no mandate to interfere in the political process of constitutional amendment.

If the duty to negotiate is worrisome, the Court’s four principles are deeply troubling. The Secession Reference suggests that each of the four can support a variety of substantive obligations that hover, invisibly, outside the Constitution. The uncertainty is compounded by the Court’s failure to provide guidelines as to the nature and scope of these obligations. At present, given the declaration that they can be abstract and general, or specific and precise, it is impossible to know or guess what those obligations may be. In addition, the Supreme Court’s list of unwritten principles is not exhaustive, and the potential for other organizing principles to provide judicial authority for the imposition of further obligations remains alive.

In conclusion, it may be useful to explore the connection between lochnering, discussed above, and the Secession Reference. Following the U.S. Supreme Court’s decision in Roe v. Wade, John Hart Ely wrote a scathing article, which became the foundation for his well-known treatise on the legitimacy of judicial review, titled Democracy and Distrust. As for Roe, he declared it a "very bad decision". In doing so he admitted reaching that conclusion, "[n]ot because [Roe] will perceptibly weaken the Court, and "not because it conflicts either with my idea of progress or what the evidence suggests is society"s". He regarded Roe v. Wade as bad, because it is "bad constitutional law, or rather because it is not constitutional law and gives almost no sense of any obligation to try to be".

Unfortunately, those remarks apply to the Supreme Court of Canada’s decision in the Secession Reference. It does not appear that the Court’s institutional credibility was undermined by the decision’s departure from accepted principles of constitutional interpretation. That said, the Court’s four principles and duty to negotiate are not constitutional law in any conventional sense, and likewise give "almost no sense of any obligation to try to be". The Secession Reference is distressing because it re-inforced two patterns that became entrenched in the Provincial Judges Reference: the first is the irrelevance of the text; and the second is the Court’s failure to heed the admonition that the text, as the supreme law, is the only true source of authority for judicial review.

Writing in the Provincial Judges Reference, Mr. Justice La Forest did not deny that unwritten rules exist or that such rules might find expression in the preamble of the Constitution Act, 1867. He hastened to add, however, that "these rules really find their origin in specific provisions of the Constitution viewed in light of our constitutional heritage". La Forest J. elaborated by explaining that "what we are concerned with is the meaning to be attached to an expression used in a constitutional provision".

Meantime, the Chief Justice’s majority opinion claimed that the Court’s interpretation of the preamble’s organizing principles was rooted in authority, including New Brunswick Broadcasting Co. v. Nova Scotia (Speaker of the House of Assembly); the doctrine of full faith and credit; the principle of federal paramountcy; the Reference re Manitoba Language Rights; and the pre-Charter concept of an implied bill of rights.

Yet it emerges, on closer examination, that most of those examples are not on point. For instance, leaving aside the question whether New Brunswick Broadcasting Co. was correctly decided, Mr. Justice La Forest persuasively demonstrated in the Provincial Judges Reference that there is no comparison, historically, between the legislature’s prerogative to control its proceedings and the status of judicial independence in a system of parliamentary supremacy. By the same token, the Manitoba Language Reference referred to the rule of law’s "implicit inclusion" in the preamble to ground an extraordinary remedial solution in principle; unlike the Provincial Judges Reference, that decision created no substantive rights or obligations. Finally, the proposition that the preamble contains implied constitutional rights that can be enforced against governments by the courts was never adopted by a majority of the Court. In short, incorporating a principle of judicial independence and then imposing institutional mechanisms on provincial legislatures exceeded all previous interpretations of the preamble. That is what prompted Mr. Justice La Forest to react against what he regarded as an unprecedented lack of respect for the constraints of text.

The Provincial Judges Reference in turn provided authority for the irrelevance of the text to be compounded in the Secession Reference. Although the former linked judicial independence to the preamble, the Court proceeded in the Secession Reference as though the text did not matter at all. There, as seen above, the judges abstracted general principles from constitutional history, promoted them as binding obligations, and then extrapolated the more specific duty to negotiate.

It could be argued that the Secession Reference is less offensive as an extension of judicial authority. Unlike the remuneration case, which resulted in the creation of salary commissions under the Constitution, the Secession Reference failed to impose immediate, concrete obligations. Even so, the unwritten principles generated by the Court there may be more dangerous in the long run. The Provincial Judges Reference articulated a particular principle to deal with the specific problem of financial security for provincial judges. Not to discount concerns about the constitutionalization of relations between the judiciary and their governments, the principles entrenched by the Court in the Secession Reference are of a different scale and magnitude. Residing, as they do, entirely outside the text, these principles have no content, and the Court is free to read them as broadly or narrowly as it pleases. This is especially true of principles that are abstract and general in nature. Uncertainty is one byproduct of this approach to interpretation; as brief reference to Lalonde v. Ontario demonstrates, the absence of limits on decisionmaking that takes place entirely outside the text is another.

In Lalonde, Ontario’s Divisional Court held that the Constitution’s protection of minority culture prevented the restructuring of Ontario’s only unilingual French-speaking hospital. The 1867 and 1982 Constitutions provide for the protection of specified minority language and education rights but do not, by their terms, confer status on unilingual minority language hospital services. Undeterred by the text, the Court declared that what was at stake was not "simply" a minority language or education issue, but a "minority culture issue". Characterizing the issue that way enabled the Court to declare that the Constitution’s "discrete notions" about language and education ought not to obscure "broader concepts", such as "the preservation of the multicultural francophone heritage of Canadians".

Lalonde did not consider it dispositive that the Constitution made definitive choices in deciding how and to what extent minority rights would be protected. Rather than respect the authority of the text, the Court inferred that, in addressing discrete aspects of the problem, the Constitution was incomplete or short-sighted. The Secession Reference provided direct authority for the proposition that the courts can fill gaps in the text, and the Divisional Court in Lalonde simply accepted that invitation. In particular, the Secession Reference suggested that specific constitutional provisions protecting minority language rights reflect a "broader principle", and added that "the protection of minority rights is itself an independent principle underlying our constitutional order". Now that unwritten principles have become the equal of, if not paramount to, the Constitution’s express provisions, the text no longer defines the scope of judicial review.

The text has been belittled and no longer represents the supreme law of the land but is, instead, subject to judicial amendment on a case to case basis. It is particularly damaging that in all three cases - the Provincial Judges Reference, the Secession Reference, and Lalonde - the judiciary’s interpretation of the Constitution served political purposes. It is difficult enough to support the view that unwritten principles can create substantive obligations, and even more difficult when such principles are invoked, outside the text, to grant judges financial security, invent a duty to negotiate, and create a right to minority culture.

Law and politics are difficult to separate, but some distinctions remain. Principled interpretations of the Constitution may be politically acceptable or not; by the same token, those that are unprincipled may also be politically acceptable or unacceptable. It should not be assumed, though, that a decision that is politically acceptable is principled as a matter of constitutional law. For the Court to forget the distinction between law and politics places its role as interpreter of the Constitution at risk. Its job is not to change the status of provincial judges , to even the political score between Ottawa and Quebec, or to decide that a unilingual hospital should not be closed. The integrity of constitutional interpretation can be a challenge to defend when the written text is involved, and is virtually impossible to support under the unwritten principles jurisprudence.

The idea that the words of the Constitution should not constrain its interpretation is not entirely new. In the Motor Vehicle Reference, Lamer J. rejected the suggestion that s.7's interpretation should be guided by the intent of its drafters: were the Court to agree, he suggested, "the rights, freedoms and values embodied in the Charter in effect become frozen in time to the moment of adoption with little or no possibility of growth, development and adjustment to changing societal needs". Resistance to the imperatives of textualism and original intent can be traced to that icon in Canadian constitutionalism, the Persons Case. There, in concluding that the word "persons" in s.24 of the Constitution includes women, Lord Sankey stated that the British North America Act, 1867 "planted in Canada a living tree capable of growth and expansion within its natural limits".

The Persons Case raised a textual question of statutory interpretation, and Lord Sankey’s living tree metaphor explicitly contemplated limits on the Constitution’s growth by interpretation. Even so, the "living tree" doctrine has over the years provided a rationale for granting the Constitution as liberal an interpretation as possible. Otherwise, as the Court suggested in the Secession Reference, the text would place the Constitution in a straitjacket. And there is little to choose between a living tree and a straitjacket.

This paper’s objective has been to flag a dangerous pattern in the Supreme Court’s interpretation of the Constitution. As seen in the discussion above, the Court has concluded that the text can be ignored whenever following it would be inconvenient. Language in the Provincial Judges Reference and the Secession Reference comes close to implying that the text is no more than a starting point. Together, these decisions maintain that whenever its provisions are deemed incomplete or misleading, the judiciary is entitled to fill in the gaps. As demonstrated throughout the paper, however, the Court’s unwritten, organizing principles are not subject to the discipline of the text and accordingly are without limits. And that is dangerous because the unwritten principles jurisprudence, on its face, appears oblivious to concerns about unlimited judicial power, and deliberately shuts out lingering doubts about the legitimacy of review. As Mr. Justice Stone pointed out many years ago, it is in the Court’s interest to be aware of those doubts and practice judicial self-restraint. In that regard it is troubling that the Supreme Court of Canada does not regard the constitutional text as a constraint on its power of review.

Nor is the choice between a living tree and a straitjacket as clearcut as the Court’s rhetoric suggests. By imposing constraints on legislatures through constitutional interpretation, judicial review can place the democratic process in a straitjacket. It is well known that constitutional amendment is not a realistic solution to Supreme Court of Canada decisions that bind governments, and the suggestion that dialogue between courts and legislatures is an answer to judicial review can only go so far. In addition, it bears noting that unwritten substantive obligations, such as minority culture rights, are not subject to the Charter’s override provision. The legitimacy of pre-empting the override through the creation of unwritten substantive obligations is open to serious question.

In the end, this paper does not reject the concept of unwritten principles. As indicated at the outset, any such position is untenable. It does suggest, however, that as a matter of constitutional interpretation, the Supreme Court’s "unwritten principles" jurisprudence is unprincipled. Except in compelling circumstances, where history, convention and practice provide corroboration, the Court should not create substantive obligations that bear no connection to the text. For all the reasons discussed above, this is not within the legitimate purview of the Court’s authority to interpret the Constitution. This is especially so when unwritten principles are invoked to support results that are transparently political.