Towards a civic republican theory of Canadian constitutional law.
Civic republican theory occupies an important place in the contemporary public law literature of some jurisdictions, but has not significantly influenced Canadian constitutional theory. (1) Moreover, and again unlike other jurisdictions, (2) there have been few theoretical accounts that provide a unified view of Canadian constitutional law rather than focusing on specific topics (i.e., particular rights) or domains (i.e., federalism or rights). (3) This essay begins to fill these gaps in the literature. I will argue that civic republicanism fits and justifies a broad range of domains of Canadian constitutional law. I build my argument on what is considered by many to be a core feature of civic republicanism, namely, the principle of non-domination, and I offer arguments that are consistent with a particular strand of civic republican theory. (4) This essay will focus that version of the theory on rule-of-law issues and on questions of individual rights. In subsequent articles, I will address the pertinence of civic republicanism for questions of federalism and group rights.
Let us begin with a sketch of civic republicanism that will be filled out by the discussion in the individual Parts of this essay. At the core of one influential modern version of civic republican theory is a concern about nondomination. (5) According to Professor Philip Pettit, it is unjustifiable for state action merely to give effect to the preferences of citizens or factions. (6) When the state acts in ways that affect citizens' priorities, civic republicanism requires that it does so on the basis of reasons which appeal to some understanding of the public good and which permit reasoned contestation of state action. (7) To permit otherwise would subject citizens to potential and actual state action that is merely an expression of the will of factions in society. Domination, according to civic republican authors, is the fact of being vulnerable to this kind of state action. (8)
For civic republicans concerned with institutional design, there are at least two ways of counteracting domination. The first is by ensuring that citizens can participate on equal terms with one another in public deliberation and the political process and that, as a result, they can define for themselves the terms of the public good under which they will be governed. The second is to create processes that deter or preclude the state from engaging in action that imposes arbitrary constraints on citizens; included among these processes is a variety of norms that require state actors to provide justifications for their actions or that prevent the state from relying on defective justifications. We will see that Canadian courts have generated doctrines that incorporate both strategies for giving effect to the civic republican norm of non-domination.
A civic republican approach to constitutional law represents an advance over a pluralist alternative. This alternative has been the main target of criticism by Professors Cass Sunstein and Frank Michelman, who have been perhaps the leading proponents of civic republicanism in the legal academy. Pluralism is also the main target of this essay. (9) In the pluralist vision, constitutional law is conceived of as an output of interest group activity. In one version, pluralism rejects the idea that judicial decision-making involves principled reason-giving. Authors claim that all such putative reason-giving is a cover for the assertion of preferences. (10) Moreover, some writers in this pluralist tradition, even as they ascribe value to deliberation and compromises forged in legislative processes, deride the institution of judicial review. (11) But judicial review is an institutional fact in Canada and any serious normative theory about Canadian constitutional law should provide arguments for structuring judicial review in ways that are conducive to legitimate processes and outcomes, rather than simply attempting to describe judicial motivations or offering wholesale criticisms of judicial review. (12) As we will see below, civic republicanism provides resources for arguments of this kind.
A second version of pluralism, which does have a prescriptive focus, argues that judicial review should aim to ensure that politically under-represented groups participate effectively in political processes. In constitutional theory, this "representation reinforcing" argument has been most influentially articulated by Professor John Hart Ely.13 Yet from a civic republican perspective, there is a significant flaw in this conception of pluralism: simply promoting participation by under-represented groups without paying close attention to the justificatory content of the claims they make or to the quality of deliberation in the political process can lead to domination as surely as does simple acquiescence to the will of the majority. In either case, mere preferences, rather than reasons related to the public good, can shape political outcomes. Furthermore, in either case, law is conceived of primarily as an instrument to put into effect specific policy preferences or objectives. (14) By contrast, as we shall see below, a civic republican approach to constitutional law focuses on the potential of legal institutions to facilitate reasoned deliberation and generates proposals that aim to realize that potential.
This essay's focus on constitutional law doctrine might be challenged by some authors working within the civic republican tradition who are critical of the institution of judicial review. One such critic, Professor Richard Bellamy, argues that judicial review undermines the position of citizens in any polity whose public institutions facilitate the equal participation of citizens in political life. (15) In his view, judicial review subjects citizens in a polity of this kind to rule by a group of individuals--judges--whose authority to issue binding decisions is not subject to deliberative processes which expose all claims to public scrutiny. Bellamy concludes that judicial review is therefore susceptible to sliding into a form of arbitrary rule and domination. (16) By contrast, other civic republicans argue that judicial review can be conducive to civic republican ends, and that the courts can function as one part of an institutional architecture that supports civic republican governance. (17) I do not intend to enter deeply into this debate, but as will be made clear in this essay, I hold to the latter position. I do so in part because this essay's objectives are interpretive and prescriptive. My posture towards civic republican critics of judicial review is the same as the one I adopt in relation to pluralist critics: I take the Canadian constitutional landscape as it is, and I attempt to explain and justify its institutions, and to offer an account that is meaningful to actors within those institutions. Wholesale criticisms of those institutions are unlikely to provide these actors with much guidance.
In Part II of this essay, I will argue that the concept of the rule of law, as it has been developed in Canada, evidences core features of civic republicanism. In Part III, I will argue that some individual rights doctrines also manifest essential characteristics of civic republican theory, including solicitude for the capacity of citizens to engage on equal terms with one another in public debates, and concern about the vulnerability of citizens to arbitrary state action.
I do not intend with this essay to claim that civic republicanism offers the only possible account of Canadian constitutional law. I aim, rather, to advance the following modest claims: (i) that civic republicanism (as it has been articulated by some scholars in law and related disciplines) offers a convincing and productive account of Canadian constitutional law; (ii) that this account is superior to the pluralist position (as that position has been advanced in law schools and political science departments); and (iii) that such an account is a contribution to the literature, given the paucity of attempts by scholars to offer overarching theories of Canadian constitutional law. The present civic republican account does not exclude others and, in particular, does not exclude a liberal alternative. Indeed, we will see that some civic republican authors, including several on whom I rely, do not view liberalism and republicanism to be incompatible. Some civic republican authors have advocated for "a liberal brand of republicanism." (18)
Despite these qualifiers, readers may be concerned that civic republicanism is indistinguishable from liberalism and may therefore believe that my reading of Canadian constitutional law is a liberal account. In order to respond to this concern, I very briefly situate this essay in a current debate about the relationship between civic republicanism and liberalism. That discussion should allay concerns about the distinctiveness of a civic republican approach. Having addressed those concerns, I will undertake the main task of this essay: a civic republican reading of Canadian constitutional law.
I. Civic republicanism and liberalism
The discussion of the relationship between civic republicanism and liberalism can productively be introduced by recounting Professor Richard Fallon's criticism of Michelman and Sunstein. According to Fallon, the authors provide neither a coherent normative program nor a clear counter-argument to liberalism. He argues that they are vague about what counts as empathetic or not self-interested deliberation. (19) This lack of precision yields two consequences, according to Fallon. First, he claims, civic republican authors make process-based arguments that purport to rest on uncontroversial premises, but in fact bring into their arguments controversial substantive premises without arguing for them. (20) Second, Fallon argues that civic republican authors do not provide a sufficiently clear account of "what rational deliberation is, how it can be practiced, and how its absence can be identified." (21) In addition, Fallon charges, Michelman and Sunstein's civic republican arguments resemble closely key aspects of liberal thought. Of the former, Fallon argues: "like many liberals, he ultimately attempts to justify contestable choices by government by appeal to the postulated outcomes of an idealized or even artificially defined reasoning process." (22) Similarly, he argues that Sunstein's argument "is avowedly 'liberal.'" He accepts, for example, that a just society would recognize rights elevated above the debates characterizing ordinary politics." (23)
A response to this litany of criticisms can begin by noting that the key distinction between liberalism and civic republicanism, in the influential modern version articulated by Pettit, is that the latter evinces a commitment to liberty as non-domination and not simply as non-interference. (24) For the purposes of constitutional law, this distinction is significant. Constitutional law scholars in the civic republican tradition will aim to offer legal prescriptions that (i) reduce the state's capacity to act in arbitrary ways, (25) (ii) conduce to effective processes of democratic deliberations which call upon the capacity of citizens "to determine the political decisions that bind them," (26) and (iii) reduce barriers to equal deliberation. (27) The main objective of these proposals is to counter domination by ensuring that state action does not merely express factional will. Broadly speaking, the proposals do so either by facilitating citizen input into government decisions or by ensuring that those decisions are justified in terms of public reasons, and not of private preferences. In response, then, to Fallon's claim that legal civic republicanism's normative program has been unfocused, I argue that the core concern for non-domination is sufficiently clear and well-defined to support substantive arguments and institutional proposals that can be progressively refined through discussion and experimentation. This essay's arguments are built on this core concern and are a first iteration of Canadian civic republican arguments and proposals.
A. Non-domination vs. non-interference
One may accept that the idea of non-domination provides a sufficiently well-defined core for the project of constructing a civic republican theory of Canadian constitutional law, yet insist that the contrast between noninterference and non-domination is not as clear as civic republicans have claimed. Professor Ian Carter has argued that liberal theories emphasize a conception of negative freedom as "absence of prevention," where "prevention can be either actual or subjunctive: one is negatively unfree to do x not only if someone has already foreclosed the option of doing x, but also if, were one to attempt to do x, someone would act so as to foreclose one's doing x." (28) Carter argues that in this formulation, negative freedom is broader than in cases in which an individual is directly coerced, and that it is these latter kinds of cases that civic republicans mistakenly and exclusively associate with the idea of negative freedom. (29) Moreover, Carter argues that although civic republicans claim that one is "unfree" in cases where one is merely vulnerable to the exercise of another's power, the better conception of freedom states that one is unfree, in the sense of having options foreclosed, only to the extent that it is probable that the holder of the power will exercise it to interfere with one's choices. (30)
In response to this criticism, Professor Quentin Skinner argues that even if the idea of negative freedom and the related concept of interference is formulated (as it has been by Carter and others) to include courses of action that individuals feel constrained to pursue or compelled to avoid, it does not capture the situation of being "condemned to living wholly at the mercy of [another's] arbitrary power." (31) Pettit similarly argues that the concept of interference, even if it is defined broadly to involve the removal of options from an agent, does not capture the republican idea of domination. According to this idea, an agent's choice can be controlled when another person has the ability to increase the probability that the agent will follow the course of action preferred by that other person. (32) In such circumstances, there is, according to Pettit, invigilation, not interference, and for a civic republican, invigilation is a form of control that undermines freedom. (33) What both Skinner and Pettit capture in their responses to their liberal critics is the insight that civic republicanism does not aim primarily to safeguard individual interests by providing bulwarks against interference from others. Rather, civic republicanism targets a particular kind of interpersonal harm, namely, that of being subject to another's will.
Pettit illustrates the distinctiveness of this form of harm by distinguishing the case in which one is subject to potential arbitrary interference from another person from the case in which one is subject to similar interference from a natural cause. If the probability of interference is identical, and one is only concerned with interference, these two cases are indistinguishable. It is only if one understands the interpersonal significance of being under the control of another that one appreciates the significance of domination. Consider the situation where the probability of interference decreases in the hypothetical case involving the natural cause. In such a situation, the threat of interference, which is the only threat of harm on the scene, also decreases. By contrast, simply decreasing the probability of arbitrary interference by another person (but not eliminating its possibility) will not affect the fact that that other person exerts control over one's choices, and that one is therefore subject to that person's will. (34) Civic republicans committed to legal institutional design aim to respond to the threat of this kind of interpersonal harm by "creating an institutional framework that promotes deliberation, [and] prevents domination by factions." (35) I will argue below that Canadian constitutional law, interpreted in a civic republican light, provides just such a framework.
B. Liberalism and civic republicanism: Distinctiveness and compatibility
It is important not to overstate the argument in the previous section. I intend with that argument only to claim that liberalism and civic republicanism are distinct to the extent that (i) the concepts of non-interference and non-domination differ from one another and (ii) the concepts are representative of significant lines of their respective traditions. In addition, from the argument thus far advanced we can conclude that those versions of liberalism that approximate the pluralist conception of democracy are also distinguishable from civic republicanism. A civic republican has good reasons for opposing versions of liberalism that conceive of politics as an instrument for advancing self-interest. She so objects because state action in this conception, as in the pluralist one, is the product of mere assertions of preferences. (36) As we have seen above, this kind of state action results in domination. So we have two reasons for claiming that civic republicanism is distinguishable from liberalism and I hope with this discussion to have convinced the skeptical reader of the pertinence of the differences highlighted. Yet although these differences are real, they should not be exaggerated. As Professor Richard Dagger has argued, liberalism and civic republicanism are distinct but not necessarily incompatible. (37) I close this Part by discussing authors who have claimed that civic republicanism and liberalism are compatible. In the next Part, we will see a particular doctrinal example of such compatibility.
There are several ways in which authors argue that civic republicanism and liberalism are compatible. Some argue that although civic republicanism and liberalism are distinguishable in many cases, application of the theories to a particular question will yield similar conclusions. (38) Others argue that one of the theories can incorporate aspects of the other. Rawls, for instance, claimed that the virtues associated with "classical republicanism" are consistent with political liberalism because the fostering of these can assist citizens and polities in preserving basic liberties. (39) Similarly, Professor Trevor Allan argues that his liberal theory of common-law constitutionalism presupposes a "wider context of a 'republican' conception of politics." (40) Still others argue for a synthesis of liberal and civic republican theory. (41) Dagger, for instance, argues that although the ideals of non-interference and non-domination are distinguishable, a concern for autonomy is common to both. He claims that we want to be free from domination because we want to be able to govern ourselves and this capacity is also threatened when our freedom is threatened by interference. (42) We will soon see that authors from each tradition offer overlapping accounts of why public justification is significant to the concept of the rule of law. Before we turn to that discussion, however, we consider the role that such justification plays in Canadian constitutional law doctrine pertaining to that concept.
II. Civic republicanism and the value of legality
Let us begin by recalling those elements of the rule of law that the Supreme Court of Canada has identified in its case law. According to the Court in the Secession Reference, the rule of law comprises the following elements: (1) "the rule of law provides that the law is supreme over the acts of both government and private persons. There is in short, one law for all"; (2) "the rule of law requires the creation and maintenance of an actual order of positive laws which preserves and embodies the more general principle of normative order"; and (3) "the exercise of all public power must find its ultimate source in a legal rule.... [T]he relationship between the state and the individual must be regulated by law." (43) In Imperial Tobacco, the Court reasons that these elements of the rule of law, as they are articulated and developed in precedent or in the express terms of the Constitution, have the force of law in Canada. (44)
The third requirement--that the relationship between the state and the individual be regulated by law--has been the basis of doctrinal development and scholarly commentary that affirms a core feature of the rule of law, namely, respect for the human agency or autonomy of citizens. (45) In the administrative law context, authors have argued that this aspect of the rule of law imposes on the state an obligation to engage the citizens it regulates in a dialogue, rather than subjecting those citizens to the state's "one-way projection of authority." (46) In Canadian administrative law, the adoption of this conception of legality by the Supreme Court of Canada has led Professor David Dyzenhaus to characterize its administrative law doctrine as evincing a "democratic view of the separation of powers." (47) According to this view, argues Dyzenhaus:
[T]he determination of the content of law is viewed in terms of a relationship of reciprocity between legislature and subject, so that interpretive authority is shared between the institutions of the legal order, including the subject who as citizen contests the law within the domain of its application to him. (48)
We will soon see that this "democratic view" resonates with a civic republican approach to the rule of law. As does Dyzenhaus' view of administrative law, such an approach requires state actors to offer public justifications for their decisions and to respect thereby the agency of citizens subject to state action. In order to articulate more clearly the relationship between this view of the rule of law and civic republicanism, I turn, perhaps surprisingly, to consider the work of an avowed liberal. That discussion will highlight the civic republican aspects of the rule of law, and will enable us to see the points of discontinuity and overlap in liberal and civic republican theories of legality.
A. Civic republicanism, legality and integrity
Professor Ronald Dworkin in Law's Empire famously argues for a concept of law-as-integrity in which the requirements of "fit" and "justification" figure prominently. In hard cases, he says, courts should aim to interpret existing constitutional materials in the best possible political or moral light, keeping in view the purposes of those materials. (49) Such a process of interpretation contributes to a "genuine political community" in which "people ... are governed by common principles, not just by rules hammered out in political compromise." (50) On this account of constitutional interpretation, citizens are participants in a society-wide process of responsive reason-giving and part of a community characterized by the mutual regard of its members. (51) In Justice in Robes, Dworkin frames these claims in terms of legality. He argues that legality requires that "the government must govern under a set of principles in principle applicable to all" (52) and notes that the process of identifying what these principles require in particular cases involves publicly defending and articulating particular conceptions of these principles. (53) The civic republican overtones of his position emerge in particularly striking manner when he frames this imperative of justification in terms of self-government. (54)
This understanding of the role of citizens in legal regulation and of the appropriate relationship of the state to citizen shares essential features with modern civic republicanism and with the above discussion of the Supreme Court of Canada's analysis of the rule of law. Consider Dworkin's requirement of principled reason-giving. The idea of law-as-integrity rejects patchwork solutions in law--solutions, that is, that cannot be defended on general grounds of principle--because such solutions, even if they have the incidental effect of increasing aggregate welfare, impose upon individuals constraints on action that cannot be defended by appeal to the public good. When the state acts in ways that affect the priorities of citizens without such justification, it denies them the benefits of membership in a Dworkinian political community and citizens denied these benefits are, in civic republican terms, subject to domination. State action in these circumstances amounts to a mere assertion of political will and does not permit rational contestation by citizens subject to it.
Despite this consonance, Dworkin rejects those theories of law that he calls communitarian, but which resemble civic republicanism. According to Dworkin, those who advance the ideal of deliberation about the public good against interest group pluralism "rely on a dubious though rarely challenged assumption: that public discussion of constitutional justice is of better quality and engages more people in the deliberative way than courts." (55) We will see in Part II that no such assumption is necessary for a civic republican theory of constitutional law, and that some writers in the civic republican tradition have expressly argued that judicial intervention in public law is necessary precisely because courts are in a better position to resolve some kinds of issues, in some circumstances, than are legislatures. (56) Before I move to that discussion, I conclude this Part by discussing the differences between a civic republican approach to the rule of law that focuses on non-domination and a liberal one that focuses on non-interference.
B. Implications of the civic republican conception of the rule of law
If Dworkin's attempt to distinguish between civic republican and liberal theories of legality does not succeed, how should we distinguish them? Dagger has drawn the distinction in a way that tracks the difference between interference and domination. The civic republican theory of the rule of law protects "from dependence on the arbitrary will of others" and stresses citizen participation in law-making. By contrast, he argues that liberal theory aims to provide "people a secure set of expectations so that they may pursue their private projects." (57) This articulation of the liberal understanding of the rule of law encapsulates the question that motivates Dworkin's argument in Law's Empire. Dworkin there takes as his central question the problem of how a theory of law can justify the state's coercive exercise of power against citizens. (58) The answer that he arrives at is law-as-integrity, and the argument leading up to that response builds on the idea of a "genuine political community." (59) These ideas have civic republican overtones, but the motivating question, with its emphasis on interference with citizens' interests, falls on the liberal side of Dagger's civic republican-liberal divide. Professor Trevor Allan similarly articulates this liberal conception and frames it explicitly in terms of non-interference when he argues that the goal of the liberal ideal of the rule of law "is to ensure that any interference with people's liberty or property is regulated by general rules, whose purposes are the promotion of some aspect of a genuine public good." (60)
This difference in conceptions of legality yields a distinction between the kinds of administrative law questions that civic republicans and liberals might address, and between their reasons for addressing certain questions. The civic republican, like the liberal, aims to shape judicial doctrine that requires agency actors to justify their decisions once a citizen's interests have been affected, but for different reasons-the liberal to increase the sphere of private freedom, the civic republican to ensure that state action does not merely reflect private preferences. The civic republican will also be deeply invested in designing institutions that promote the direct participation of interested parties in administrative governance. (61) The civic republican conception of the rule of law can be expressed in institutional designs that enable citizens to pre-empt domination by actively participating in public policy processes. Professor Henry Richardson's writing on civic republicanism provides a helpful starting point for such a program of institutional design.
Richardson argues that policy-making often involves a process of giving progressively more specific content to vague formulations of policy ends. (62) According to him, there are typically a variety of means of giving effect to a given policy end, and policy makers, in the process of narrowing the range of possible means, narrow the possible interpretations of that end. Because some interpretations will be revealed after shared reflection or common experience to be not reasonably achievable, policy makers will abandon them. In turn, as the policy ends become more sharply defined, certain policy means will similarly be left by the wayside. (63) A civic republican will seek to ensure that citizens and the associations that represent them participate in the collective project of defining the public ends that are set out in generally applicable statutes, and of selecting the specific means to achieve those ends. (64)
Canadian regulation-making already incorporates mechanisms that promote this kind of deliberation and public involvement, and a civic republican project of institutional design would seek to expand the extent of public involvement. (65) For instance, the Statutory Instruments Act includes deliberation-forcing mechanisms which seek to ensure that regulations--at a minimum--do not impose unjustified constraints on citizens. The ex ante constraints include the mandatory examination of regulations by the clerk of the Privy Council, in consultation with the Deputy Minister of Justice. Any proposed regulation is examined to determine whether it satisfies legal requirements, whether it is consistent with standard regulatory practices and whether it is judiciously drafted. (66) If, after examination, there are any concerns about these issues, they are brought to the attention of the regulation-making body. (67) If this process of examination has not occurred, the clerk of the Privy Council may refuse to register and certify the proposed regulation, and can thereby prevent it from coming into force. (68) These processes seek to ensure that regulations are subject to close expert scrutiny in light of concerns relating to the public good. And once a regulation has been registered, its publication in the Canada Gazette opens it to public examination. (69) A civic republican concerned with the possibility of domination would seek to enable all those directly affected by proposed regulations opportunities to participate in processes of public examination. This kind of participation functions as a safeguard against domination because it limits the influence of private interests on the regulation-making process. (70) Such consultation does not primarily aim to protect citizen interests against unjustified state interference, although this may be an effect. These forms of civic republican consultation rather aim to involve citizens in constituting the rules by which they are governed, and to reduce thereby the risk of domination by factions who have captured the regulatory process. (71)
From the discussion thus far we have seen that the liberal and republican conceptions of the rule of law are distinguishable but not necessarily incompatible. The conceptions are distinguishable because liberals envision the rule of law as protecting citizens against interference, while civic republicans conceive it as protecting against domination. We have also seen that for both traditions the rule of law requires that state action be based on principled and public reasons, and that the traditions are in this respect compatible. More importantly, given the overarching purposes of this essay, we have seen that the civic republican account of the rule of law doctrine counters the pluralist view of law. The civic republican reading, which reflects the state of positive law in Canada, requires that state action be justified in terms of the public good and that processes of public participation empower affected parties to deliberate about the regulations that affect them. The civic republican view of legality aims to pre-empt the influence of mere preferences on state action. Whether expressed in judicial decisions or in the institutions of regulation-making, the rule of law so understood aims at the goal of non-domination, and therefore contradicts the pluralist view.
Now that we have seen general arguments in favour of a civic republican account of legality, we can turn our attention to Canadian constitutional law doctrine concerning individual rights. In the next Part, I will argue that significant portions of that body of doctrine can be understood in light of two constitutional principles identified by contemporary writers on civic republicanism: the principle against arbitrary state action and that in favour of the political equality of citizens. (72) I will begin by showing how these principles are present in the case law before arguing that these two principles reflect the distinctively civic republican concern about non-domination.
III. Charter protections and civic republican values
It is in the equality and freedom-of-expression jurisprudence of the Supreme Court of Canada that we can see most clearly constitutional protections for the political equality of citizens and safeguards against arbitrary state action. Consider first the equality jurisprudence of the Court. The Court examines equality claims using two broad criteria: first, whether the challenged government action perpetuates a constitutionally protected group's pre-existing disadvantage and experience of prejudice, and second, whether the action imposes a disadvantage based on stereotypes. (73) The first criterion is directly concerned with the political equality of citizens. The fact of a group's historical disadvantage suggests that its members are unable to engage on equal terms with other citizens in the political process. (74) The judicial focus on stereotypes in the equality context indicates an additional concern about arbitrary state action. The Court has reasoned that unwarranted legislative distinctions which are based on constitutionally protected immutable characteristics violate the equality interests of citizens who bear those characteristics. Such distinctions impose on those citizens arbitrary constraints. (75) Because stereotypes are presumed to be inaccurate, any policy that is justified on the basis of a stereotype will be arbitrary in the sense that it does not track the relevant interests of the claimant and the claimant's group. (76) By contrast, where legislative distinctions are tailored to the actual experiences of citizens who possess such characteristics, even if these characteristics are constitutionally protected, the Court has found that no discrimination results, precisely because such distinctions do not impose arbitrary constraints on citizens. (77)
Canadian freedom-of-expression doctrine similarly evinces a dual concern for the political equality of citizens and about the possibility of arbitrary state action. In its hate-speech jurisprudence, the Court has noted that such speech has the potential to exclude vulnerable groups from public discourse. (78) This element of hate-speech jurisprudence reflects a desire to protect those groups' ability to participate as equals in the public sphere. A similar desire to protect the political equality of women underwrites the Court's freedom-of-expression jurisprudence in the area of pornography. (79) Legislative restrictions on pornography are constitutionally justified, according to the Court, at least in part because they protect the position of women as political equals. The public dissemination of degrading pornography, the Court reasons, undermines that position.
This concern for political equality is also the express rationale for permitting legislative limits on campaign spending: absent such limits, some segments of society will dominate the political process. (80) In addition, the Court has expressed a concern about the potential for hate speech to lead to arbitrary state action. The Court has reasoned that the marketplace of ideas rationale for freedom of expression, which one might invoke to argue for the protection of viewpoints expressed in hate speech, underestimates the capacity of such speech to motivate irrational and arbitrary political outcomes. (81) In its freedom-of-expression jurisprudence, the Court permits state regulation of hate speech at least in part as a means of guarding against such outcomes.
Generally, when legislative or executive action has been found to infringe a constitutionally-protected right, courts will require that that infringement be justified under section 1 of the Canadian Charter of Rights and Freedoms. (82) The analysis under that section requires the government to show that an impugned measure has a legitimate objective, that the means chosen to advance that objective are rationally connected to that objective, that the infringement is minimally impairing of the claimant's rights and that any infringement is proportional to the objectives sought. (83) Perhaps most pertinent for present purposes, when government action has no public-regarding justification--when for instance, it manifests the coarsest forms of partisan preference--it will automatically fail constitutional scrutiny. (84) Because such state action affects constitutionally significant individual interests and does so without a plausible public justification, it is not subject to the assessment of consequences that is the purpose of much of the section 1 analysis.
A. Harms of domination and not (only) harms of interference
This necessarily incomplete and brief overview of rights doctrines serves to illustrate the general claim that a civic republican interpretation of the jurisprudence is an advance over a pluralist alternative. On their face, the doctrines are designed to ensure that the state does not give effect to mere assertions of political will or preferences, and in some instances, the doctrines permit the state to counter such assertions. For those readers concerned that this civic republican reading is indistinguishable from a liberal interpretation, it is perhaps helpful to set out a reading of the doctrine that focuses on the principle of non-interference. With this liberal version articulated, I will sharpen the relationship between the concept of non-domination and the principles in favour of equal treatment and against arbitrary state action.
One plausible interpretation of the Court's section 15 doctrine concerning historically disadvantaged groups would claim that the reasoning aims to prevent unjustified interference in the affairs of those individuals most likely to be subjected to such interference. (85) A similar interpretation of the Court's concern for politically vulnerable victims of hate speech and pornography is possible: under conditions of social inequality, forms of expression that denigrate already vulnerable groups may threaten the security of these groups' members. (86) Finally, arguments grounded in the rationale of non-interference can be made about the doctrines concerning stereotypes and illegitimate governmental objectives. A liberal might object to the failure to justify governmental actions in these cases because such interference reduces the sphere in which individuals can pursue their interests. In part, argues the liberal, this interference arises from the ability of dominant groups to dictate to minority groups the terms by which the latter understand themselves. (87)
Because as we have seen above, liberalism and civic republicanism are compatible traditions, there is no reason to think that these liberal readings of the doctrines exclude civic republican possibilities. A civic republican interpretation may acknowledge that the doctrines achieve the effect of protecting citizens against interference, but it may also claim that they aim to empower citizens to contribute actively to public discourse and to ensure that state action is grounded in public reasons. These measures aim to prevent the state from acting in ways that merely evince the preferences of factions. They aim to protect citizens against domination. These civic republican objectives are most evident in the electoral campaign cases, where reasonable constraints on campaign spending are found to be constitutional because they limit the ability of some interests to dominate political discourse. Such dominance has the potential to yield governments and government policy that reflect the resource-backed preferences of factions in society, rather than the deliberative judgments of the polity.
The goal of non-domination can also be seen in the Court's concern for the position in the public sphere of minority groups subject to hate speech and for women in a society where public displays of pornography are permitted. These forms of expression can have the effect of excluding some groups from public discourse, and such exclusion ultimately degrades the quality of reasoning in the political process and threatens to subject these groups to domination. This concern about reasoned deliberation may arise even in the absence of any direct interference with the interests of members of minority groups or women. Even if there is no causal link between pornography and violence against women or between hate speech and violence done to minority groups, and therefore not interference, a civic republican may object to these kinds of speech because they send the message that women and minority groups are not full and equal participants in the political sphere. (88) Of course, this message can be understood as interference with those groups, because, for instance, it undermines the ability of their members to fashion freely their own identities. (89) Nonetheless, I maintain that there is a distinctive civic republican harm present in these cases: the impugned speech undermines reasoned public decision-making by excluding affected citizens from deliberation, and therefore increases the potential for domination of these citizens by the state. (90) There is, therefore, a compelling civic republican interest in permitting the state to restrict such expression. (91)
The civic republican concern for non-domination is also evident in the Court's equality jurisprudence. Consider first that element of the case law which addresses the vulnerability of historically disadvantaged groups to majority predation. In cases where these groups are denied equal benefit of the law, we have good reasons to worry that their members were not able participate as equals in public decision-making, and that the resulting state action failed to include reasoned consideration of their perspective. The risk in such cases is that the government action, at least with respect to the relevant group, represents a mere assertion of political will to which that group's members are vulnerable. Here again, the civic republican might recognize that the state has interfered with the priorities of citizens, but she would also point to the distinctively republican harm of domination.
Consider finally the significance of stereotypes to a civic republican-influenced reading of section 15. At one level, the significance is relatively straightforward. Because stereotypes represent invalid generalizations, they are arbitrary and subject those who are the objects of state action which reflects stereotypes to an expression of political will, not reasoned political judgment, and therefore to domination. Yet, as Moreau has argued, not all unjustified generalizations are constitutionally problematic. (92) She focuses on the various ways in which government action that relies on or perpetuates stereotypes based on the section 15 grounds restricts the autonomy of individuals who are the objects of such action and she claims that stereotypes pertaining to those grounds are more likely than other kinds of generalizations to restrict citizens' autonomy. (93) I suggest that in addition to having these autonomy-restricting effects, we presume that it is particularly unlikely that state action that generalizes on the basis of some of the constitutionally protected grounds will match the relevant interests of the citizens affected, and that therefore such state action reflects an arbitrary assertion of political will.
Because of historical experience, we in general assume that one's race, nationality, ethnic origin, colour, religion, and sex are irrelevant to legislative or regulatory distinctions. (94) State action that relies on generalizations related to these grounds is presumed to be arbitrary. For some of the protected grounds, and for some subsets of the protected grounds, this presumption about arbitrariness is joined by another about political inequality. For instance, we presume that those who belong to racialized minority groups with little political clout will struggle to have their interests adequately represented in the political process. (95) We further presume that inaccurate legislative or regulatory generalizations about these groups will reflect this lack of representation, and will not adequately take into account their perspectives. (96) For the civic republican, then, stereotypes related to the constitutionally protected grounds are in general unlikely to represent reasoned judgments, and when the grounds overlap with political disadvantage, we further presume that the stereotypes reflect a failure of political representation. In either case, the civic republican conceives of government action resulting from stereotypes to entail flawed reasoning and presumptively to deny citizens a justification for that action. Such government action renders those who are the objects of the stereotypes vulnerable to domination. (97)
B. Responses to objections and a defense of under-enforcement
One might object that the overview of constitutional doctrines just undertaken does not demonstrate a strong link between Canadian public law and civic republicanism. If civic republicanism is concerned with domination, as the objector would note, Canadian constitutional doctrines are obviously under-protective. The objector might point to the constitutional provision that most obviously protects citizens against arbitrary state action, namely section 7 of the Charter. A violation of section 7 occurs when state action infringes a citizen's interest in life, liberty or security of the person in a way that violates a principle of fundamental justice and one principle of fundamental justice is that state action cannot be arbitrary. The objector might ask us to note that the protection against arbitrariness, in the section 7 context, only arises when one of the constitutionally protected interests--life, liberty or security of the person--is at play. (98) Similarly, the objection might continue, the constitutional protection against arbitrary state action in the context of equality rights does not extend to all vulnerable groups and does not even extend to groups that are clearly vulnerable, such as the economically disadvantaged. (99) Yet even if we concede that constitutional law protections are generally under-inclusive, this concession does not represent a serious challenge to a civic republican theory of Canadian constitutional law. To see why, we should recall the nature of the interest protected by civic republicanism, and consider the various institutions in which that interest may be defended.
The judicial protection of only some constitutional rights serves a distinctly civic republican purpose: it creates constitutional room for institutions that are in general better able than courts to facilitate the ability of citizens to engage in self-directed activity and to develop their own resources to check the state's tendencies to engage in dominating behaviour. To understand this claim, we will need to consider two related sets of constitutional theory: the decision-rules literature and the writing on judicial under-enforcement of constitutional norms. In the decision-rules literature, a distinction is drawn between constitutional meaning and constitutional doctrine. (100) Writers note that even when, by drawing upon the full range of interpretive techniques available in constitutional law, we can discern with relative ease the meaning of a constitutional provision, that meaning cannot be directly enforced by the courts. In enforcing a constitution, courts generate and apply doctrinal rules that have embedded within themselves a set of judgments about the effects of doctrine and the institutional capacities of courts. (101)
Judicial reasons that under-enforce a constitutional norm fall within the set of reasons that reflect a gap between the constitution's meaning and its doctrinal interpretation. (102) Such under-enforcement may result from concerns about the relative capacity of courts to enforce or to determine the contents of rights. For instance, a court may accept that a constitutional right to equality or to protection of one's life against arbitrary state action potentially imposes a positive obligation on the state to provide social assistance to individuals. Such a court may, nonetheless, decline to offer a judicial interpretation of the constitution that either announces that right or specifies its content; such a court would also, as a consequence, decline to formulate a judicial doctrine that enforces that right. The justification for such restraint lies in institutional considerations. A court may reasonably decide that it lacks the capacity to make the relevant determinations about what constitutes an infringement of a right to minimum social welfare and would, in light of that understanding of its limited capacity, leave the task of defining such a right to the legislature and of enforcing it to the executive. (103) For the judiciary to function effectively as a protector of citizen autonomy, it should fend off threats to its credibility. (104) A judiciary that consistently rendered judgments in areas beyond its institutional capacities would lose credibility in the eyes of citizens, as its reasons would lack justifications that would enable them to be viewed as legitimate. (105)
These institutional reasons for restraint are joined by more robust concerns about the autonomy or agency interests of citizens. Autonomy in the civic republican sense implies that citizens are capable of participating in the political process, that the political process solicits such participation and that citizens are treated as self-originating sources of claims. (106) These aspects of civic republican autonomy require democratic rule and imply that citizen participation in defining the norms that govern them is important to that form of rule. (107) Democratic institutions may fail to give effect to these norms, but this fact only points to the need for institutions to assist in the project of realizing these norms. The judiciary is one such institution, but because of the complexity and scale of the work of democratic governance, the dominant share of it falls to the elected institutions of the state. Citizen input into these institutions enables the state to function in ways that track the relevant interests of citizens but at the same time, these institutions are a significant source of constraints on citizen autonomy. Judicial review, when it rules out state action that clearly and unjustifiably undermines citizen autonomy, can act as a supplement to well-functioning democratic institutions. However, because courts are not broadly representative and are institutionally incompetent to manage complex social, political and economic matters, judicial review cannot take the place of legislative and executive branches as the primary authoritative locus of citizen deliberation about the public good and of safeguards against infringements of citizen autonomy. (108) Courts acting with restraint do so in recognition of this fact.
One may object that these arguments about judicial interpretation speak only to the choice of courts not to expand interpretations of the constitutional text beyond some limits, but do not address the under-inclusiveness of the text itself. To respond to this challenge, we should assess the reasons for the generally parsimonious nature of constitutional language. (109) For a constitution to function effectively as a framework for interactions between the state and citizens, its provisions should be relatively few and general. (110) Moreover, open-ended constitutional language provides space for citizens and legislatures to offer interpretations of a constitution which may be more expansive than those offered by the judiciary. Such differences in the scope of interpretation may in part be attributable to courts' recognition of their own institutional limits, and to their strategic decision to articulate and enforce constitutional rights only in those contexts where their interventions can most effectively promote political equality and safeguard citizens from arbitrary state action. More importantly, the potential for interpretations of the constitution on issues which are more expansive in scope than those that can reasonably be addressed by the judiciary gives to the members of a democratic polity the opportunity to exercise and develop their own deliberative capacities as they engage in political debate about the meaning of the Constitution. Far from undermining the case for civic republicanism, the limited number and general nature of the rights protected by the express constitutional text and in judicial decisions reinforce the civic republican rationale for constitutional rights. They are the preconditions for democratic deliberation and for effective judicial interventions. (111)
In this Part, I have (i) argued that civic republicanism fits and justifies significant portions of Charter doctrine, and (ii) defended that argument against several objections. The treatment of the issues was not intended to be exhaustive, but I hope to have provided a reasonably fair treatment of the relevant literatures, and to have shown the applicability of a civic republican approach to some Charter debates. My core claim has been that existing doctrine reflects a commitment to civic republicanism and refutes pluralist claims about Canadian constitutional law. I have also argued that my account is distinguishable from a liberal interpretation of the doctrine. Constraints of space preclude me from applying the civic republican approach to open doctrinal questions in Canadian constitutional law. I can only suggest that courts seeking civic republican resolutions of such questions will balance considerations of institutional competence and democratic capacity in light of the general prohibition against domination. (112)
In this essay, I have presented the case for a civic republican theory of Canadian constitutional law in the domains of rule of law and Charter rights jurisprudence. The theoretical model against which I have argued is that of interest group pluralism and I have, in the course of developing the argument, distinguished civic republican arguments from liberal ones. As I conclude the essay, I respond to two final pluralist objections and suggest future directions for research. The first pluralist objection states that I have taken too seriously what the courts have said about constitutional reasoning. This reasoning, says the pluralist critic, is merely cover for the preferences of the judges, and to pretend that the reasoning, rather than preferences, is what determines the outcomes of decisions is to participate in this charade. My initial response is that even if judges in constitutional cases are motivated by policy preferences, they are obliged to justify their interpretive choices with reasons that are available publicly and must be defended against similarly reasoned challenges. From the civic republican perspective set out in this essay, such an obligation legitimates constitutional judgments and they are illegitimate to the extent that they do not satisfy this obligation. (113)
More generally, I claim that the pluralist critique does not accurately represent the stakes at issue in constitutional interpretation. Any adequate account of state action requires arguments that are sensitive to the normative particularities of state action. (114) Professor Jon Elster has framed pluralist insensitivity to these particularities as evidencing confusion between the logic of the market and that of the public forum. In the marketplace, the only question to be answered is: how are my preferences to be satisfied? But the public forum is driven by an entirely different set of questions. (115) In the forum, we are concerned not only with our own wants, but with how decisions affect others and with questions of justice. In the forum, it is not enough to aggregate private preferences. Rather, citizens there deliberate together about what the public good is and about what it requires of the state and of citizens. (116)
A final pluralist critic might claim that I have fundamentally misunderstood the point of constitutional interpretation, and in particular, Charter interpretation. Such interpretation, the critic argues, properly aims not at the civic republican goals I have set out, but at a substantive outcome. The true purpose of constitutional interpretation, the critic claims, is to improve the material conditions of those who bear the brunt of social, political and economic inequalities in our society. The civic republican focus on public justification and non-domination, says the critic, obscures this fundamental purpose and potentially offers a patina of legitimacy to outcomes that are at variance with the substantive goal.
My response to this criticism begins with a claim about the nature of constitutions. In my view, constitutions instantiate diverse purposes that in turn reflect the diversity of their societies. Some substantive positions fall outside the reasonable range of opinion, but for those that do not, constitutions can function as the meeting grounds for ongoing deliberation and discussion about what we value, about how we can accommodate those who hold views different from our own, and about how they can accommodate us. Such a view of the constitution acknowledges the partiality of any particular view of constitutional values and the fallibility of all who hold such views. (117) In my opinion, a civic republican approach to constitutional law better respects this capacity of constitutions to provoke deliberation about how citizens in a society should live together than does a view of the constitution that imagines law to be an instrument to achieve specific policy outcomes. (118) To this pluralist critic, then, I say that a civic republican approach to constitutional law does address issues of inequality in society, because inequality is a fundamental constitutional concern of our polity, but does so within a wider analysis of the problem of domination and of the questions of justification and deliberation to which that problem gives rise.
I will in future essays develop this publication's civic republican theory as I analyze the domains of federalism and group rights. I shall claim that viewing those domains through the lens of civic republicanism will yield arguments that counter purely instrumentalist views of Canadian constitutional law. I hope to show that the principle of non-domination provides a unifying theme for Canadian constitutional law, and that we can understand our political institutions as primarily providing occasions for the exercise and respect of the political agency and deliberative capacities of citizens.
(1) Some scholars have lamented the paucity of theoretical work in Canadian constitutional law. See Robert Howse & Sujit Choudhry, "Constitutional Theory and the Quebec Secession Reference" (2000) 13 Can JL & Jur 143 at 144-45. The authors cite a long list of authors who have done theoretical work in constitutional law, but none explicitly writes in the civic republican tradition. By contrast, civic republicanism has influenced legal theory in a variety of other jurisdictions, most notably the United States. For a survey, see Samantha Besson & Jose Luis Marti, "Law and Republicanism: Mapping the Issues" in Samantha Besson & Jose Luis Marti, eds, Legal Republicanism: National and International Perspectives (Oxford: Oxford University Press, 2009) 3.
(2) For examples of ambitious and wide-ranging theoretical accounts of constitutional law in other jurisdictions, see e.g. Michael Dorf & Charles Sabel, "A Constitution of Democratic Experimentalism" (1998) 98 Colum L Rev 267 and Richard Bellamy, Political Constitutionalism: A Republican Defense of the Constitutionality of Democracy (Cambridge: Cambridge University Press, 2007).
(3) One notable, monograph-length exception is David Beatty, Constitutional Law in Theory and Practice (Toronto: University of Toronto Press, 1995). I assume that such comprehensive accounts are valuable because they create the possibility for constitutional interpretation and its attendant legality-related benefits (see below, Part II).
(4) There are of course many strains of civic republican thought, some of which may not accord with arguments I make in this essay. I do not purport to resolve these internal debates, and instead make claims about the conceptions of civic republicanism that are limited in the ways described in the main text.
(5) On this conception of domination, see Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 1997) at 56: "The acts of interference perpetrated by the state must be triggered by the shared interests of those affected under an interpretation of what those interests require that is shared, at least at the procedural level, by those affected ... Every interest and every idea that guides the action of a state must be open to challenge from every corner of the society." See also Cass R Sunstein, "Beyond the Republican Revival" (1988) 97 Yale LJ 1539 at 1548-49: "The purpose of politics is not to aggregate private preferences, or to achieve an equilibrium among contending social forces. The republican belief in deliberation counsels political actors to achieve a measure of critical distance from prevailing desires and practices, subjecting these desires and practices to scrutiny and review."
(6) Pettit, ibid at 204-05.
(7) For the distinction between modern and classical theories of civic republicanism, see Sunstein, supra note 5 at 1539-40. See also Frank Michelman, "Law's Republic" (1988) 97 Yale LJ 1493 at 1495. In this essay, I will adopt the modern usage.
(8) "What is required for non-arbitrary state power ... is that the power be exercised in a way that tracks, not the power-holder's personal welfare or world-view, but rather the welfare and worldview of the public." Pettit, supra note 5 at 56. See also Michelman, supra note 7 at 1504-05: "Republicanism has been, par excellence, the strain in constitutional thought that has been sensitive to both the dependence of good politics on social and economic conditions capable of sustaining 'an informed and active citizenry that would not permit its government either to exploit or dominate one part of society or to become its instrument,' and the dependence of such conditions, in turn, on the legal order" [footnotes omitted].
(9) For this contrast between pluralism and civic republicanism, see Sunstein, supra note 5 at 1542-45. See also Michelman, supra note 7 at 1503.
(10) For instance, Professor Rainer Knopff writes of a series of judicial decisions on Charter issues, which he claims are drafted in value-neutral language: "non-judgmental neutrality in intense moral conflicts really represents the victory of one of the extremes. The apparent neutrality is just a disguised way of saying tu perds" See Rainer Knopff, "Courts Don't Make Good Compromises" in Paul Howe & Peter Russell, eds, Judicial Power and Canadian Democracy (Montreal: McGill-Queen's University Press, 2001) 87 at 90. For an overview of the Canadian literature, see James B Kelly, Governing with the Charter: Legislative and Judicial Activism and Framers' Intent (Vancouver: UBC Press, 2005), ch 1.
(11) Knopff, ibid at 92-93.
(12) Some critics of judicial review seem to alternate between wholesale condemnations of judicial review and more specific critiques of judicial doctrine. Compare, for example, Professors Morton and Knopff's critique of the Supreme Court of Canada's interpretive choices in chapter 2 of their important text and their claims about the effects of judicial review on the practices of democratic persuasion in chapter 7 of that same work: FL Morton & Rainer Knopff, The Charter Revolution and the Court Party (Peterborough, Ont: Broadview Press, 2000). The present essay avoids this ambiguity and aims to focus the discussion on interpretive questions, rather than on the larger debate about the legitimacy of judicial review. In addition to the criticism of simple attitudinalist models of judicial review in the main text are others, including those which focus on its inattentiveness to institutional constraints and influences on judges. See e.g. Cornell W Clayton & Howard Gillman, eds, Supreme Court Decision-Making: New Institutionalist Approaches (Chicago: University of Chicago Press, 1999).
(13) John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Mass: Harvard University Press, 1980). For the contrast between Ely's position and that of modern civic republicans, see Michael A Fitts, "Look Before You Leap: Some Cautionary Notes on Civic Republicanism" (1987-88) 97 Yale LJ 1651 at 1651. For a similar contrast, drawn between majority and constitutional rule, see Christopher L Eisgruber, ed, Constitutional Self-Government (Cambridge, Mass: Harvard University Press, 2001) at 59-62.
(14) For this criticism of pluralism, see Sunstein, supra note 5 at n 63: "There is some connection between the republican belief in political equality and some pluralist efforts to promote access to the political process, see, e.g., J Ely, Democracy and Distrust (1980), but the pluralist view places little premium on deliberation or citizenship." For an example of an approach to Canadian constitutional law that conceives of law as an instrument to achieve progressive social ends, and that sharply criticizes outcomes at variance with these, see e.g. Allen C Hutchinson, "Condition Critical: The Constitution and Health Care" in Colleen M Flood, Kent Roach & Lorne Sossin, eds, Access to Care, Access to Justice: The Legal Debate Over Private Health Insurance in Canada (Toronto: University of Toronto Press, 2005) 101 at 103-105.
(15) Bellamy, supra note 2 at 220-21, argues: "So long as a system of equal votes, majority rule and party competition-however interpreted-offers a plausible system for giving citizens an equal say in the ways collective arrangements are organised-including those of the democratic process-then a self-constituting democratic constitution that avoids dominating through arbitrary rule will have been secured."
(16) Ibid at 151.
(17) For a defense of judicial review as an instrument of facilitating democratic deliberation, see Iseult Honohan, "Republicans, Rights and Constitutions: Is Judicial Review Compatible with Republican Self-Government?" in Besson & Marti, supra note 1, 83 at 89: "Freedom requires a strong institutional structure of accountability and transparency within which even democratic government exercises its power. This does not undermine the principle of collective self-government, but means that the way that self-government is assured through the ensemble of government institutions."
(18) Besson & Marti, supra note 1 at 9, and citations therein. For liberal republicanism, see Sunstein, supra note 5 and Dagger, infra note 36.
(19) Richard H Fallon, "What Is Republicanism, and Is It Worth Reviving?" (1989) 102 Harv L Rev 1695 at 1728-29, 1732.
(20) Ibid at 1729.
(21) Ibid at 1733.
(22) Ibid at 1730.
(23) Ibid at 1731.
(24) Pettit, supra note 5 at 8-9, writes: "liberalism has been associated over the two hundred years of its development, and in most of its influential varieties, with the negative conception of freedom as the absence of interference, and with the assumption that there is nothing inherently oppressive about some people having dominating power over others, provided they do not exercise that power and are not likely to exercise it." Of course, any attempt to find essential aspects of traditions as broad and inclusive as liberalism and republicanism is fraught with difficulties. In this essay, I focus tightly on a particular debate and accept the characterizations of the traditions offered by the participants in that debate. This strategy has the benefit of focusing the inquiry and freeing me from constantly qualifying my claims when I speak of civic republicanism and liberalism. I emphatically do not intend to capture all strands of the two traditions in this essay.
(25) See Pettit, supra note 5.
(26) See e.g. Besson & Marti, supra note 1 at 20.
(27) Professor James Bohman frames this demand of equality in terms of the rule of law. He writes: "[t]he rule of law demands ... that each person can bind others and be bound by them; that is, each must possess not just one-way, but two-way normative powers, powers to create and accept obligations second-personally." James Bohman, "Cosmopolitan Republicanism and the Rule of Law" in Besson & Marti, supra note 1, 60 at 64.
(28) Ian Carter, "How are Power and Unfreedom Related?" in Cecile Laborde & John Maynor, eds, Republicanism and Political Theory (Oxford: Blackwell Publishing, 2008) 58 at 67 [footnotes omitted].
(29) Ibid at 68.
(30) Ibid at 70.
(31) Quentin Skinner, "Freedom as the Absence of Arbitrary Power" in Laborde & Maynor, eds., supra note 28, 83 at 99.
(32) Philip Pettit, "Republican Freedom: Three Axioms, Four Theorems" in Laborde & Maynor, eds., supra note 28, 102 at 112.
(33) For the various ways in which an agent can be controlled, without interference, see ibid at 113.
(34) Ibid at 124.
(35) See Martin Loughlin, "Towards a Republican Revival?" (2006) 26 Oxford J Legal Stud 425 at 428-29.
(36) Richard Dagger, Civic Virtues: Rights, Citizenship, and Republican Liberalism (Oxford: Oxford University Press, 1997) at 105. Dagger argues that interest group pluralism "is consistent ... with that strand of liberalism, running from Hobbes through Bentham and James Mill." Ibid.
(37) Ibid at 12.
(38) See e.g. Will Kymlicka, "Liberal Egalitarianism and Civic Republicanism: Friends or Enemies?" in Anita L Allen & Milton C Regan Jr, eds, Debating Democracy's Discontent: Essays on American Politics, Law, and Public Philosophy (Oxford: Oxford University Press, 1998) 131 at 146.
(39) John Rawls, Political Liberalism (New York: Columbia University Press, 2005) at 205.
(40) TRS Allan, Constitutional Justice: A Liberal Theory of the Rule of Law (Oxford: Oxford University Press, 2001) at 24.
(41) See e.g. Dagger, supra note 36 and Henry S Richardson, Democratic Autonomy: Public Reasoning About the Ends of Policy (Oxford: Oxford University Press, 2002).
(42) Richard Dagger, "Republican Virtue, Liberal Freedom, and the Problem of Civic Service" (Paper presented at the annual conference of the American Political Science Association, Boston, 2001), online: Department of Philosophy, Penn State University <http://philosophy.la.psu.edu/jchristman/autonomy/Dagger.PDF>. Dagger also argues for a reconciliation of civic republican and liberal values when he gives further detail to the notion of autonomy. In order to make this claim, he sets out what he views to be the distinguishing values of the two traditions. According to him, "liberals place the greatest value upon individual rights and personal autonomy, classical republicans upon civic virtue and public responsibility." Dagger, supra note 36 at 12. He begins his attempt at reconciling these two sets of values when he notes that a citizen is free, in the republican sense, if she participates as a member of her community in its governance. Dagger claims that the position of a citizen is a role and that to fulfill that role faithfully one must further the civic good. It is in this sense that to participate in government is to engage in self-government. Ibid at 15. Finally, says Dagger, no one can be autonomous through individual effort alone and one owes obligations to the wider polity that assists one in becoming autonomous. Those obligations are discharged through duties that evidence one's civic virtue. Ibid at 17.
(43) Reference re Secession of Quebec,  2 SCR 217 at para 71, 161 DLR (4th) 385 [footnotes omitted].
(44) British Columbia v Imperial Tobacco Canada Ltd, 2005 SCC 49,  2 SCR 473 at paras 58-60.
(45) For a similar argument linking Fuller's conception of legality with the interests of citizens in exercising their moral agency, see Allan, supra note 40 at 58: "it [the rule of law, as conceived by Fuller] provides the means for co-operation between citizens, enabling all to further their own interests within the constraints of justice."
(46) The expression is Fuller's. See Lon L Fuller, The Morality of Law, revised ed (New Haven, Conn: Yale University Press, 1969) at 221. For the facilitative aspect of this concept of the legality, see ibid at 223. For the claim that the exercise of administrative discretion should be conceived of as a "dialogue," see Genevieve Cartier, "Administrative Discretion as Dialogue: A Response to John Willis (Or: From Theology to Secularization)" (2005) 55 UTLJ 629 at 630 and citations therein. For the claim that rule-making should be similarly conceived, see Genevieve Cartier, "Procedural Fairness in Legislative Functions: The End of Judicial Abstinence" (2003) 55 UTLJ 217 at 241: "The best way to put a check on arbitrariness is through increased participation of the individuals affected by the decisions and a true responsiveness on the part of decision makers."
(47) See e.g. David Dyzenhaus, "Constituting the Rule of Law: Fundamental Values in AdministrativeLaw" (2002) 27 Queen's LJ 445. See also David Dyzenhaus & Evan Fox-Decent, "Rethinking the Process/Substance Distinction: Baker v. Canada" (2001) 51 UTLJ 193.
(48) Dyzenhaus, ibid at 501.
(49) Ronald Dworkin, Law's Empire (Cambridge, Mass: Belknap Press, 1986) at 90, ch 6. The above paragraph draws from Hoi Kong, "The Forms and Limits of Federalism Doctrine" (2008) 13 Rev Const Stud 241 at 243-44.
(50) Ibid at 211.
(51) Ibid at 211-15.
(52) Ronald Dworkin, Justice in Robes (Cambridge, Mass: Belknap Press, 2006) at 176. Professor Frederick Schauer has noted that because of the generality of reasons, the act of reason-giving implies a commitment to making future decisions that fall within the scope of those reasons' application. See Frederick Schauer, "Giving Reasons" (1995) 47 Stan L Rev 633 at 644. This feature of reason-giving sees intrinsic value in consistency and further restricts the capacity of those who participate in a reason-giving enterprise to act in self-serving ways. Ibid at 653.
(53) Dworkin, ibid at 184.
(54) Says Dworkin, "The root idea we are now exploring-0that individual freedom is furthered by collective self-government-assumes that the members of a political community can appropriately regard themselves as partners in a joint venture...." Ronald Dworkin, Freedom's Law: The Moral Reading of the American Constitution (Oxford: Oxford University Press, 1996) at 25. If Dworkin is understood in this way, there is no necessary conflict between his version of legality and a version that is appropriate to the administrative context. For the claim that there is such conflict, see David Dyzenhaus, "Law as Justification: Etienne Mureinik's Conception of Legal Culture" (1998) 14 SAJHR 11 at 24-27. If the essential insight of Dworkin's theory of legality is that the state must offer public justifications for its actions, nothing prevents a court from assessing whether another branch of government has offered such a justification and deferring to that justification, when appropriate. For the claim that Dworkin has provided just such an argument, see TRS Allan, "Dworkin and Dicey: The Rule of Law as Integrity" (1988) 8 Oxford J Legal Stud 266 at 273-76. In particular, Allan writes: "a Dworkinian version [of the Rule of Law need not] be committed to the rejection, or curtailment, of official discretion: but the notion of public law rights provides a legitimate device for constraining its exercise." Ibid at 276.
(55) On the general overlap between Dworkin's work and civic republicanism generally, see Besson & Marti, supra note 1 at 34. For an articulation of this aspect of civic republicanism, in writing on constitutional law, see Michelman, supra note 7 at 1514-15.
(56) See Michelman, supra note 7. See also Cass R Sunstein, After the Rights Revolution: Reconceiving the Regulatory State (Cambridge, Mass: Harvard University Press, 1990) at 38-45, for the argument that citizens may be prevented from pursuing or constructing preferences in ways that require regulatory intervention to overcome.
(57) Dagger, supra note 36 at 61.
(58) Dworkin, supra note 49 at 190.
(59) Ibid at 211.
(60) Allan, supra note 40 at 39.
(61) For a civic republican defense of the administrative state, which argues that administrative agencies, because of their expertise and delegated authority, are better positioned to engage in public reasoning than legislatures or courts, see Mark Seidenfeld, "A Civic Republican Justification for the Bureaucratic State" (1992) 105 Harv L Rev 1512 at 1542. He writes: "Administrative agencies, however, fall between the extremes of the politically over-responsive Congress and the over-insulated courts. Agencies are therefore prime candidates to institute a civic republican model of policymaking." One can contrast this republican focus on administrative design with liberal theorists' focus on courts as privileged sites for rational deliberation about fundamental moral questions. For this liberal focus, see Allan, supra note 40 at 85 and Ronald Dworkin, A Matter of Principle (Oxford: Oxford University Press, 1985) at 68-71.
(62) Richardson, supra note 41 at 104-105.
(63) Ibid at 123.
(64) Ibid at 180: "Modes of feedback and reflective oversight, in which the public and the legislature remain aware of and involved to a degree in more detailed policy-making, are essential to the whole process counting as one of popular will-formation."
(65) On the importance of such deliberation to the civic republican conceptions of liberty, see Besson & Marti, supra note 1 at 17: "a republican can endorse a complex conception of liberty that embodies both a private and a public dimension, and even emphasize the importance of effective political participation, as Habermas does...."
(66) Statutory Instruments Act, RSC 1985, c S-22. Section 3(1) requires that any regulation-making authority shall cause to be forwarded to the Clerk of the Privy Council three copies of the regulation, and section 3(2) provides that
the Clerk of the Privy Council, in consultation with the Deputy Minister of Justice, shall examine the proposed regulation to ensure that :
(a) it is authorized by the statute pursuant to which it is to be made;
(b) it does not constitute an unusual or unexpected use of the authority pursuant to which it is to be made;
(c) it does not trespass unduly on existing rights and freedoms and is not, in any case, inconsistent with the purposes and provisions of the Canadian Charter of Rights and Freedoms and the Canadian Bill of 'Rights; and
(d) the form and draftsmanship of the proposed regulation are in accordance with established standards.
(67) Ibid, s 3(3). In addition, if a regulation-making body is in doubt as to whether a proposed statutory instrument is a regulation, it must forward a copy to the Deputy Minister of Justice who will determine whether, if made, the instrument would be a regulation. Ibid, s 4.
(68) A proposed regulation, unless it is exempted by the Governor in Council (s 20(a)), comes into force only if it is registered by the clerk of the Privy Council. Ibid, s 9(1).
(69) Section 11(1) requires publication of registered regulation in the official gazette of Canada. If, however, a regulation is not published, then, subject to some conditions, no individual can be convicted of an offence under that regulation. Ibid, s 11(2).
(70) For criticisms of civic republican models of administrative law design, see Steven P Croley, "Theories of Regulation: Incorporating the Administrative Process" (1998) 98 Colum L Rev 1 at 81-85 and Jonathan R Macey, "The Missing Element in the Republican Revival" (1988) 97 Yale LJ 1673.
(71) Some jurisdictions have begun to experiment with electronic means of encouraging participatory regulation-making. See e.g. Cary Coglianese, "E-Rulemaking: Information Technology and the Regulatory Process" (2004) 56 Admin L Rev 353.
72 See e.g. Richardson, supra note 41 at 27. Sunstein similarly argues that across a range of American constitutional and administrative law doctrines, legislatures and administrative agencies are under an obligation to provide public-regarding reasons for their actions and as such are precluded from simply giving effect to private preferences that are unconnected to such reasons, and from acting in ways that undermine the fundamental dignity of citizens. Sunstein, supra note 56, ch 2.
(73) R v Kapp, 2008 SCC 41,  2 SCR 483 at paras 18-24 [Kapp].
(74) The Supreme Court of Canada in Egan v Canada,  2 SCR 513 at 520, 124 DLR (4th) 609 specifically invoked the language of "discrete and insular minorities" when discussing the kinds of groups eligible for constitutional protection under the Constitution's equality provision. A civic republican focus on political equality is particularly attentive to circumstances in which there is a deliberative failure which causes the perspectives of some groups to be overlooked. For an overview of kinds of deliberative failure, see Charles F Sabel & William H Simon, "Destabilization Rights: How Public Litigation Succeeds" (2004) 117 Harv L Rev 1016 at 1064-65. For a specifically civic republican focus on this function of equality, see Besson & Marti, supra note 1 at 20: "The kind of political equality propounded by republicanism should lead to the effective equal capacity of every citizen to influence or determine the political decisions that bind them."
(75) See Vriend v Alberta,  1 SCR 493, 156 DLR (4th) 385 [Vriend]; and refer to this decision also for general arguments about the nature of arbitrary state action, as seen through the lens of section 15. See also Trociuk v. British Columbia (Attorney General), 2003 SCC 34,  1 SCR 835 at para 24; and Corbiere v Canada (Minister of Indian and Northern Affairs),  2 SCR 203 at para 13, 173 DLR (4th) 1.
(76) For this conception of arbitrary power, see Pettit, supra note 5 at 55. For the claim that human dignity is affirmed where arbitrary obstacles to the flourishing of individual autonomy are removed, see generally Denise G Reaume, "Discrimination and Dignity" (2003) 63 La L Rev 645 at 673.
(77) See e.g. Gosselin v Quebec (Attorney General), 2002 SCC 84,  4 SCR 429 at para 56 [Gosselin]: "The legislator is entitled to proceed on informed general assumptions without running afoul of s 15, ... provided these assumptions are not based on arbitrar y a nd demeaning stereotypes" [footnotes omitted]. I am aware that there is a distinction drawn in the literature between the idea of discrimination and the concept of an infringement of a right to equality. For present purposes, I use the term "discrimination" as a short-hand for any infringement of the right to equality. For a fuller discussion of the correspondence factor, see section C below.
(78) R v Keegstra,  3 SCR 697 at 746, 61 CCC (3d) 1 [Keegstra].
(79) R v Butler,  1 SCR 452 at 479, 89 DLR (4th) 449 [Butler].
(80) Libman v Quebec (Attorney General),  3 SCR 569 at para 47, 151 DLR (4th) 385. See also Harper v Canada, 2004 SCC 33,  1 SCR 827.
(81) Keegstra, supra note 78. For an argument associating the goals identified in the main text with civic republican values, see Besson & Marti, supra note 1 at 21-22.
(82) Part I of the Constitution Act, 1982, being schedule B to the Canada Act 1982 (UK), 1982, c 11.
(83) These requirements constitute the Oakes test. See R v Oakes,  1 SCR 103, 26 DLR (4th) 200.
(84) For cases in which the Supreme Court of Canada found legislation unconstitutional because the legislative objective was objectionable, see R v Big M Drug Mart Ltd,  1 SCR 295, 18 DLR (4th) 321; Vriend, supra note 75.
(85) See anti-subordination theories of equality. For the classic statement, see Ruth Colker, "Antisubordination Above All: Sex, Race, and Equal Protection" (1986) 61 NYUL Rev 1003 at 1005, 1007.
(86) For concerns about these kinds of harms, see Keegstra, supra note 78 at 747; and Butler, supra note 79 at 503.
(87) See Sophia R Moreau, "The Wrongs of Unequal Treatment" (2004) 54 UTLJ 291 at 299: "Someone who has been defined a benefit on the basis of a stereotype has been publicly defined by another group's image of him. Rather than being allowed to present himself and his circumstances as he understands them, he has been presented in the manner of another's choosing."
(88) See the risk of "harm to society at large" considered in Butler, supra note 79.
(89) Moreau, supra note 87.
(90) In this respect, the civic republican analysis resembles that of scholars who conceive of freedom of expression in terms of the "distribution of communicative resources." For the latter, see Richard Moon, The Constitutional Protection of Freedom of Expression (Toronto: University of Toronto Press, 2000) at 7.
(91) See Professor Alexander Meiklejohn's arguments in Alexander Meiklejohn, Free Speech and Its Relation to Self-Government (Clark, NJ: Lawbook Exchange, 2000).
(92) Moreau, supra note 87 at 298.
(93) Ibid at 299.
(94) For an assessment of one part of that history, see Bruce Ryder, "Racism and the Constitution: The Constitutional Fate of British Columbia Anti-Asian Immigration Legislation, 1884-1909" (1991) 29 Osgoode Hall LJ 619. For an extended treatment of this presumption and its effects on American constitutional doctrine, see Robert Post, "Prejudicial Appearances: The Logic of American Antidiscrimination Law" (2000) 88 Cal L Rev 1.
(95) See e.g. Patricia Hughes, "Recognizing Substantive Equality as a Foundational Constitutional Principle" (1999) 22 Dal LJ 5. For a similar argument about the social construction of disability by majority institutions, see Dianne Pothier, "Miles to Go: Some Personal Reflections on the Social Construction of Disability" (1992) 14 Dal LJ 526.
(96) For an analysis of these considerations, in the American context, see Reva Siegel, "Why Equal Protection No Longer Protects: The Evolving Forms of Status-Enforcing State Action" (1997) 49 Stan L Rev 1111.
(97) It is important to note that a government may create a regulatory regime that fails to correspond to the actual circumstances of a constitutionally protected group yet still not engage in stereotyping. For instance, when a minor lack of fit is only and evidently the result of unavoidable difficulties of regulatory design, it would be unreasonable for a claimant or anyone else to perceive that the claimant's group has been stereotyped.
(98) For an analysis of these interests and the principle prohibiting arbitrary state action, see Chaoulli v Quebec (Attorney General), 2005 SCC 35,  1 SCR 791 at paras 131, 133. For the assessment of the related question of when a statutory delegation of authority will be found to be so broad as to permit arbitrary action, see David J Mullan, Administrative Law (Toronto: Irwin Law, 2001) at 126 -27.
(99) Gosselin, supra note 77 at para 35. For the possibility that section 7 may be extended to impose a positive obligation on states to provide social assistance, see ibid at paras 80-83.
(100) Mitchell N Berman, "Constitutional Decision Rules" (2004) 90 Va L Rev 1 at 3. See also Richard H Fallon Jr, Implementing the Constitution (Cambridge, Mass: Harvard University Press, 2001) and Kermit Roosevelt III, The Myth of Judicial Activism: Making Sense of Supreme Court Decisions (New Haven, Conn: Yale University Press, 2006).
(101) Berman, ibid.
(102) For the general idea of under-enforcement, see Lawrence G Sager, Justice in Plainclothes: A Theory of American Constitutional Practice (New Haven, Conn: Yale University Press, 2004). For a civic republican argument highlighting the significance of regulating the relationships between the branches of government to the project of protecting against domination, see Richardson, supra note 41 at 9.
(103) See on this point Sager, ibid at 87: "Some principles of constitutional justice are wrapped in complex choices of strategy and responsibility that are properly the responsibility of popular political institutions."
(104) For the relationship between judicial opinions and public opinion, see Barry Friedman, "The Politics of Judicial Review"(2005) 84 Tex L Rev 257.
(105) For this concern about credibility costs, see Jesse H Choper, Judicial Review and the National Political Process: A Functional Reconsideration of the Role of the Supreme Court (Chicago: University of Chicago Press, 1980) at 201-02, 258. For a consideration of how judicial reasons may fail to be perceived to be legitimate, see Herbert Wechsler, "Toward Neutral Principles of Constitutional Law" (1959) 73 Harv L Rev 1.
(106) Richardson, supra note 41 at 63.
(107) For a civic republican defense of populism, see ibid at 72.
(108) See Larry D Kramer, The People Themselves: Constitutionalism and Judicial Review (New York: Oxford University Press, 2004), for a stronger version of this claim that enables the legislature to override unpopular judicial decisions. See also Mark Tushnet, Taking the Constitution Away from the Courts (Princeton: Princeton University Press, 1999).
(109) I say "generally," because some provisions of a constitution can be quite specific. Included among these are age requirements and amending formulas. In the main text, I am speaking to those provisions of a constitution that provide broader, more open-ended normative guidance, such as those provisions governing the separation of powers, federalism or the rights of citizens. For a similar distinction, see Tushnet, ibid at 9-14, contrasting "thick" provisions which contain "a lot of detailed provisions describing how the government is to be organized" (at 9) with "thin" provisions which "[give] us the opportunity to construct an attractive narrative of American aspiration" (at 12).
(110) On this point about the relationship between democratic deliberation and constitutional parsimony, see Sager, supra note 102 at 141: "A constitution can significantly enhance political judgment over the concerns of justice only if it restricts itself to demands so basic and so durable that they can generally and reasonably function as dominant and nonnegotiable." For the idea of a constitution as framework for legal interaction, see Cass R Sunstein, Designing Democracy: What Constitutions Do (Oxford: Oxford University Press, 2001) at 6: "the central goal of a constitution is to create the preconditions for a well-functioning democratic order, one in which citizens are genuinely able to govern themselves."
(111) For a summary of arguments claiming that aggressive judicial review enervates a constitutional polity and undermines the ability of the elected branches to interpret the Constitution, see Tushnet, supra note 108 at 57-65 and, in the Canadian context, see Richard Sigurdson, "The Left Legal Critique of the Charter: A Critical Assessment" (1993) 12 Windsor YB Access Just 117.
(112) For an analysis that considers issues of institutional competence, and of the complexity and scope of constitutional issues when one adopts a non-liberal approach to constitutional adjudication, see Margot Young, "Unequal to the Task: 'Kapp'ing the Substantive Potential of Section 15" (2010) 50 Sup Ct L Rev (2nd) 183.
(113) For this quality of judicial decision-making see Christopher L Eisgruber, Constitutional SelfGovernment (Cambridge, Mass: Harvard University Press, 2001) at 59-62.
(114) Jon Elster, "The Market and the Forum: Three Varieties of Political Theory" in James Bohman & William Rehg, eds, Deliberative Democracy: Essays on Reason and Politics (Cambridge, Mass: MIT Press, 1997) 3.
(115) Ibid at 10-11.
(116) Ibid at 24-25.
(117) I draw this fallibilistic approach to constitutional law from the pragmatic theory of Peirce and Dewey. For a discussion of this theory, see Hilary Putnam, Pragmatism: An Open Question (Cambridge, Mass: Blackwell Publishing, 1995) at 21. For the civic republican significance of compromise, under conditions of reasonable disagreement, see Bellamy, supra note 2.
(118) For a general argument about instrumentalism in legal theory and its negative effects on the conception of legality, see Brian Z Tamanaha, Law as a Means to an End: Threat to the Rule of Law (New York: Cambridge University Press, 2006).
Hoi Kong, Assistant Professor, Faculty of Law, McGill University. I thank Eric Adams, Robert Leckey, Dwight Newman, Mark Walters, and Gregoire Webber for written comments that greatly strengthened this essay's arguments. I also thank participants at workshops at the Faculty of Law of Queen's University and of McGill University for probing questions that helped sharpen this essay's focus. I am grateful to the Borden Ladner Gervais Research Fellowship Program for funds that were instrumental in bringing this essay to conclusion, and to Adela Maciejewski Scheer and Owen Ripley for their assistance with the references in this essay. Finally, I thank the article's two anonymous reviewers for pressing me to clarify the relationship between civic republicanism and liberalism. All errors are my own.