Back to basics: a critical look at the Irwin Toy framework for freedom of expression.Just over twenty years ago, in Irwin Toy Ltd v Quebec (AG) (1) the Supreme Court of Canada constructed a special analytical framework for the resolution of freedom-of-expression cases under s 2(b) of the Charter. (2) That analytical framework both identified the questions that courts are required to answer when called upon to determine whether governmental action challenged on the basis of s 2(b) violates the right to freedom of expression, and prescribed the principles that are to be applied when each of these questions is being answered.
In setting out that framework, the Supreme Court provided us with its understanding of two of the building blocks of any coherent theory of freedom of expression as a constitutional right-(1) the meaning to be given to freedom of expression, and (2) the manner in which one determines whether governmental action infringes upon freedom of expression so defined. Of these two building blocks, the first is clearly the more important, since it determines the range of interests that s 2(b) protects. However, the second is a necessary and far from unimportant component of any such theory, for without it one has no legal basis upon which to decide in a given case whether or not the impugned governmental action adversely affects one of the protected interests in a manner, or to a degree, that should engage the concern of the courts.
The Irwin Toy framework was harshly criticized by a number of scholars and counsel in the years immediately following its adoption, and for a broad range of reasons. (3) To this point, however, that criticism has fallen on deaf ears; in fact, the Court appears to have ignored it entirely. (4) The Court has used the framework consistently in resolving s 2(b) cases since it was first articulated and, as its recent judgment in Ontario (Public Safety and Security) v Criminal Lawyers' Association3 makes clear, continues to invoke it today.
The fact that the framework is still being used in spite of the criticism levelled against it years ago might suggest that it can now be said to have stood the test of time, and that it should be accepted uncritically as an established part of our s 2(b) jurisprudence. I think that would be a mistake. Like the scholars and counsel who subjected the framework to critical scrutiny following its initial adoption, I consider the framework to be seriously flawed, and there is no less reason now to identify its flaws and urge the Court to correct them than there was when it first emerged. Not only is it important to the quality of our constitutional jurisprudence over the longer term that the Court's approach to s 2(b) issues be grounded in sound understandings of the basic elements of freedom of expression analysis, it is also important to the credibility of the Supreme Court as the ultimate arbiter of such issues.
My concerns with the Irwin Toy framework can be easily summarized: (1) the meaning the Court has given to freedom of expression for the purposes of s 2(b) lacks a solid justiicatory basis, ignores general interpretive principles the Court has adopted in relation to the Charter, encourages pointless and wasteful litigation, and fails to appreciate the symbolic function that an instrument like the Charter performs; and (2) the roadmap the Court prescribes for determining whether or not governmental action infringes on freedom of expression is inconsistent with the Court's own prior jurisprudence, lacks logical coherence, misapplies a feature of American free speech jurisprudence of questionable merit, and is incomplete. (6)
This paper is divided into four parts. In the first, which is very short, I set out the basic elements of the Irwin Toy framework. In the second, I subject to critical scrutiny the meaning of freedom of expression that that framework embodies, and in the third I do the same with the approach to finding an infringement on freedom of expression that it embodies. I then conclude with some observations about recent jurisprudential developments under the Charter generally and s 2(b) in particular that suggest that at least some members of the Supreme Court might be willing to reconsider one or two aspects of the framework and place the Court's understanding of these two building blocks of its approach to freedom of expression on a sounder footing.
A. The Irwin Toy framework
The Supreme Court developed its analytical framework for freedom of expression in a systematic and careful manner over the course of several pages in Irwin Toy. It began with the understanding of the meaning of freedom of expression that it had decided to adopt for the purposes of s 2(b) and then turned to its understanding of the manner in which governmental action was to be scrutinized to determine whether or not it infringed on an activity protected by that meaning. Various aspects of the reasoning it employed in the course of developing each of these understandings will be considered below when I subject the understandings to critical scrutiny. At this stage, it is sufficient to set out the basic elements of those understandings. That is a task made much easier than it might otherwise have been by the fact that the Court conveniently summarized the entire framework in the following brief passage, which appears at the end of the first part of its reasons for judgment:
When faced with an alleged violation of the guarantee of freedom of expression, the first step in the analysis is to determine whether the plaintiff's activity falls within the sphere of conduct protected by the guarantee. Activity which (1) does not convey or attempt, and thus has no content of expression or (2) which conveys a meaning but through a violent form of expression, is not within the protected sphere of conduct. If the activity falls within the protected sphere of conduct, the second step in the analysis is to determine whether the purpose or effect of the government action in issue was to restrict freedom of expression. If the government has aimed to control attempts to convey a meaning either by directly restricting the content of expression or by restricting a form of expression tied to content, its purpose trenches upon the guarantee. Where, on the other hand, it aims only to control the physical consequences of particular conduct, its purpose does not trench upon the guarantee.... If the government's purpose was not to restrict free expression, the plaintiff can still claim that the effect of the government's action was to restrict her expression. To make this claim, the plaintiff must at least identify the meaning conveyed and how it relates to the pursuit of truth, participation in the community, or individual self-fulfilment and human flourishing. (7)
Careful readers will have noticed that the meaning of freedom of expression is provided in the first two sentences of this passage, (8) while the measure of infringing governmental action takes up the rest. They will also have noticed that, as part of that rest, the Court invokes the three rationales or purposes that, in Irwin Toy and other s 2(b) cases, (9) it has said underlie our collective decision to include freedom of expression as one of the fundamental rights protected by the Charter. (10)
Before I begin my critical analysis of the Irwin Toy framework, I should make it clear that this framework was not designed for, and therefore should not be taken to deal with, all of the different kinds of cases that s 2(b) has shown itself to be capable of generating. It was designed for what I will call the standard or typical s 2(b) case, one in which-as in Irwin Toy itself-the claim is made that a government has restricted the ability of people to express themselves in the manner they wished to. It was not designed for cases involving claims of access to public (or for that matter private) property for expressive purposes, (11) or for cases involving claims that s 2(b) imposes a positive obligation on a government to act, (12) or for cases involving claims that a government has compelled someone to express themselves against their wishes. (13) The Court has developed distinct approaches for each of these other kinds of s 2(b) cases. I will not be assessing the strengths and weaknesses of those approaches in this paper. I am concerned only with the strengths and weaknesses of the basic framework the Court has established to deal with standard or typical s 2(b) cases.
I should also make it clear that this framework says nothing about the application of s 1 to freedom-of-expression cases. Except tangentially, I will therefore not be addressing the role that that provision of the Charter has played in s 2(b) cases, important as that role has been. In fact, it is fair to say that, precisely because of the Irwin Toy framework, and particularly the extremely broad meaning it gives to freedom of expression, most s 2(b) cases are resolved at the s 1 stage.
B. The meaning of freedom of expression
The definition of freedom of expression that emerges from Irwin Toy is any non-violent activity that conveys or attempts to convey a meaning. That definition has been held to include expressive activities as diverse as commercial advertising to children, (14) the publication by the media of the details of matrimonial proceedings, (15) communicating for the purposes of prostitution, (16) threatening violence, (17) the expression of hate messages,18 the production and distribution of obscene materials, (19) the publication of falsehoods, (20) the casting of a ballot in a referendum, (21) the production and possession of child pornography, (22) and running in a school board election. (23) It has also been suggested in obiter by one member of the Court, and implied by others, that it catches as well such expressive activities as committing perjury, defrauding by deliberate misrepresentation, counselling suicide, conspiring to commit a crime, counselling the commission of a crime and a host of other similar kinds of expression that Parliament has for a long time prohibited in the Criminal Code. (24)
This definition of freedom of expression has two noteworthy aspects to it: (1) the fact that no expressive activity is excluded on the ground of the content of the message being conveyed; and (2) the fact that violent expressive activities are excluded. The choice of a definition that contained these two features was a deliberate one and the Court in Irwin Toy provided a rationale for each. The decision not to deny protection on the basis of content was said to flow from the conviction that freedom of expression was entrenched in the Charter "to ensure that everyone can manifest their thoughts, opinions, beliefs, indeed all expressions of the heart and mind, however unpopular, distasteful or contrary to the mainstream." (25) The "diversity of ideas and opinions" thereby generated was said to be of benefit to "a free, pluralistic and democratic society" because of its "inherent value both to the community and to the individual." (26) Reference was made to Justice Cardozo's famous assertion that freedom of expression is "the matrix, the indispensable condition of nearly every other form of freedom," (27) and to Justice Rand's description of it as "little less vital to man's mind and spirit than breathing is to his physical existence." (28) Also invoked was a passage from the European Court of Human Rights' decision in the Handyside case that emphasized the connection between the "pluralism, tolerance and broadmindedness without which there is no 'democratic society'" and the need to protect "not only ... 'information' or 'ideas' that are favourably received or regarded as inoffensive or as a matter of indiference, but also ... those that offend, shock or disturb the State or any sector of the population." (29)
The decision to exclude violent expressive activities was grounded in the distinction the Court made in Irwin Toy between the content and the form of expressive activity. All expressive activity necessitates the use of some form or method of communicating and, as the Court put it, "[t]he content of expression can be conveyed through an infinite variety of forms of expression: for example, the written or spoken word, the arts, and even physical gestures or acts." (30) But, the Court continued, while s 2(b) protects "all content of expression," it does not protect all forms. In particular, it does not protect "violence as a form of expression." (31) "It is clear," the Court said, "that a murderer or rapist cannot invoke freedom of expression in justification of the form of expression he has chosen." (32) To support this assertion, the Court relied upon its earlier decision in Dolphin Delivery, in which Justice McIntyre had said that while labour picketing was protected by s 2(b), such protection would not extend to "acts of violence." (33) The Court in Irwin Toy found additional support for excluding violent expression in the proposition-which it simply asserted-that "freedom of expression ensures that we can convey our thoughts and feelings in non-violent ways without fear of censure." (34)
I turn now to my critique of these two features of the meaning given to freedom of expression in Irwin Toy, beginning with the second, the exclusion of violent expression.
1. The exclusion of violent expression
I agree with the Court's view that violent forms of expression should not be given protection under s 2(b). My complaint here is therefore not with the Court's position on that substantive issue. It is, rather, with the poor job the Court did in Irwin Toy of justifying that exclusion. In fact, I think it is fair to say that the Court provided no real justification for it at all. (35) The mere fact that violent expressive activities had been excluded by Justice McIntyre in Dolphin Delivery counts for very little, since he provided no reasons for excluding them in that case. And the proposition that freedom of expression makes it possible for us "to convey our thoughts and feelings in non-violent ways without fear of censure" provides no meaningful rationale for it either. That the guarantee of freedom of expression enables us to express ourselves in non-violent ways tells us nothing about whether or not we should be denied the benefit of its protection if we choose to express ourselves in violent ways.
In two subsequent cases, the Court has proffered additional justifications for denying protection to violent expression, but those justifications are also open to criticism. In Keegstra, decided a year after Irwin Toy, Chief Justice Dickson referred to "the extreme repugnance of [violent expression] to free expression values." (36) However, the nature of that repugnance was not explained. In the lengthiest justification yet, given in the City of Montreal case, which was decided in 2005, Chief Justice McLachlin and Justice Deschamps suggested that
[v]iolent expression may be a means of political expression and may serve to enhance the self-fulfilment of the perpetrator. However, it is not protected by s 2(b) because violent means and methods undermine the values that s 2(b) seeks to protect. Violence prevents dialogue rather than fostering it. Violence prevents the self-fulfilment of the victim rather than enhancing it. And violence stands in the way of finding the truth rather than furthering it. (37)
I am not convinced by that justification either. Violence can be said to prevent dialogue in the obvious sense that the victim of the violent activity may be so badly injured as to be unable to respond to his or her attacker, or perhaps even killed. Violence can also have a chilling effect on the willingness of both its immediate victims and at least some of the people who become aware of it-for example, members of the same racial minority as the victims of a racially motivated assault--to engage in expressive activities in the future. (38) But one has only to imagine a political assassination to realize that some violent acts can in fact spawn a great deal of dialogue amongst the general public--about the reasons for it, about the appropriate governmental reaction to it, and so on. It is true that the victims of the violence may not be able to participate in that dialogue, but there has never been any suggestion that, in order to engage the self-government rationale, the immediate targets of the expressive activity in question have to participate in the discussions and debates to which that expressive activity can be said to give rise. The fact that violence might prevent rather than enhance the self-fulillment of the victim does not mean that that rationale is undermined, given that, as Chief Justice McLachlin and Justice Deschamps concede, that same violence "may serve to enhance the self-fulillment of the perpetrator." (39) It has never been suggested anywhere else in the Court's s 2(b) jurisprudence that, in order for that rationale to be engaged, the expressive activity in question has to further the self-fulfillment of both the speaker and the listener(s). On the contrary, the Court has shown itself to be willing to protect commercial expression by corporations in cases like Ford (40) in spite of the fact that, as artificial legal entities, corporations are incapable of self-fulfillment; according to the Court, such expression deserves protection because it enhances the self-fulfillment of those who hear or read it in their capacity as consumers. And finally, violence need not stand in the way of finding truth--it is not impossible that significant positive political and social change could come out of a political assassination, at least over the longer term.
Richard Moon has proffered an alternative justification for excluding violent expression to that provided by the Court, but I am not persuaded by it either. His justification is premised on the acceptance of what he calls a relational conception of freedom of expression, which he prefers to the individualistic conception that in his view underlies the Court's approach. A relational approach, he says, entails seeing "[t]he value of expression [as resting] on the social nature of individuals and the constitutive character of public discourse." (41) If that is the true value of expression, he contends, violent expression should be denied protection because it "is an abuse of the communicative process. It attacks in the most basic way the relationship of communication." (42) The problem I have with this line of argument is that it risks denying protection to much more than violent expression. If the gravamen of the test for exclusion is whether or not the expression in question "attacks ... the relationship of communication," one runs the risk that any expressive activity that can be said to adversely affect the "communicative process" would be excluded from the reach of s 2(b). Laws governing defamation, pornography and hate speech might have to be held to be beyond the reach of s 2(b), and hence immune from challenge thereunder, if such a test were used, no matter how draconian those laws were.
Is there a more compelling justification for denying constitutional protection to violent expressive activities than the Court and Richard Moon have provided? I believe there is. In my view, the best justification is one grounded in an understanding of the rationales that are said to underlie freedom of expression that stems from the conviction that, in this country, democratic self-government, the search for truth and individual self-realization are not only best pursued but also must be pursued in non-violent ways, at least if the expressive activities said to further one or more of those goals is to receive constitutional protection. In other words, our conceptions of democratic self-government, the search for truth and individual self-realization, as constitutional values, include the conviction that we are to engage in the pursuits that further them in non-violent ways.
Those conceptions can be defended on a number of grounds. They are consistent with the admonition expressed by the Court in R v Big M Drug Mart that "it is important not to overshoot the actual purpose of the right or freedom in question, but to recall that the Charter was not enacted in a vacuum, and must therefore ... be placed in its proper linguistic, philosophic and historical contexts." (43) They are also consistent with the language used in s 2(c) of the Charter to describe the kind of freedom of assembly we have been granted: it is freedom of peaceful assembly. The presence of that important qualifier means that the rationales for that fundamental freedom would have to be understood as coming with a non-violence rider. Given the close relationship between the two rights within the Charter, not only in terms of their proximity in the text but also in functional terms, it would be anomalous not to read a similar rider into freedom of expression. These conceptions also reflect the fact that s 7 of the Charter protects the rights to "life" and "security of the person," both of which are clearly threatened by violent conduct. Finally, conceiving of those three rationales in a manner that includes such a rider can be said to inhere in our commitment to the rule of law, a principle that is implicitly recognized in the preamble to the Constitution Act, 1867 (44) and explicitly invoked in the preamble to the Charter. The rule of law is generally understood to constrain the exercise of power on the part of the executive branch of government. However, at a deeper level, the rule of law can be said to imply a rejection of the use of force as a way of resolving disputes within society, and a concomitant preference for resolving them through the agency of the law. And that preference holds true whether the disputes be political (democratic self-government), philosophical (search for truth) or personal (self-realization).
I should acknowledge that two of the people who have commented critically on the Irwin Toy framework have taken issue with the Court's exclusion of violent expression from the purview of s 2(b). Jamie Cameron has done so on the ground that such an exclusion ignores what she sees as an important structural feature of the Charter--the presence of s 1, the reasonable limits provision. (45) In her view, the existence of s 1 requires that courts exercising the power of judicial review under the Charter postpone consideration of interests that might be invoked in support of the impugned governmental action--in this context, primarily life and security of the person-until they reach that provision. I am not persuaded by that criticism, and for two reasons. The first is that the Charter contains a number of rights and freedoms that either explicitly or implicitly require courts to consider such competing interests within the confines of the rights and freedoms themselves. The Legal Rights category provides numerous examples of this--the right to be secure against unreasonable search and seizure granted by s 8 is the most obvious-but one can also find examples in some of the other categories. (46) One cannot plausibly contend, therefore, that there is a hard-and-fast rule that competing interests can only be factored into the analysis, and balanced against the right or freedom at stake, within the context of s 1. The second reason is based on the conceptions of the rationales for protecting freedom of expression for which I have contended above. If one accepts those conceptions, and if one takes seriously, as I believe one should, the implications of adopting a purposive approach to the interpretation of Charter rights and freedoms, then it is only those expressive activities that can fairly be said to further one or more of the rationales for protecting freedom of expression that are entitled to the protection that s 2(b) affords. That means that, in the case of violent forms of expression, there is no need to balance interests such as life and security of the person against the interest in freedom of expression; the latter interest--at least as a constitutional interest--is simply not engaged.
Dale Gibson's criticism of the Court's exclusion of violent forms of expression in Irwin Toy was very different from that of Jamie Cameron. His concern was that, while some such forms of expression can safely be denied protection under s 2(b), excluding all of them would result in the scope of freedom of expression being "seriously curtailed." (47) To avoid that result, he argued that the Court should adopt an approach that calls for an inquiry in each case into whether or not violence constituted the pith and substance of the expressive activity in question. (48) Only if an affirmative answer to that question is given should protection be denied; otherwise protection should be granted, and the case should be resolved at the s 1 stage. I am not persuaded by this critique either. While a blanket exclusion of violent forms of expression will curtail the reach of s 2(b) to some degree, it will surely be to a very limited degree. Moreover, that limited curtailing will, for the reasons I have given above, be fully justified. Finally, I have doubts about the viability of the adapted pith and substance test that he proposes. The American First Amendment experience with the "speech" (protected) versus "conduct" (not protected) distinction gives us little reason to be optimistic that a test based on a judicial assessment of which of two plausible ways of characterizing an activity is the better one will lead to principled results. (49)
I move now to the other and, in my view, more important feature of the Court's definition of freedom of expression in Irwin Toy--its decision not to deny protection to any expressive activity on the basis of the content of the message being conveyed.
2. The protection of all content
I have two closely related but nevertheless analytically distinct concerns with this feature of the definition, and will address each in turn. The first is that it allows for the possibility that expressive content that furthers none of the rationales for including freedom of expression in the Charter will receive constitutional protection; the second is that it extends the reach of s 2(b) too far.
Before I develop the first of these concerns, I should make it clear that, while I disagree with the Court's decision to give protection to all expressive content, I believe that s 2(b) should be interpreted broadly. In fact, there are very few content-defined categories of expression that I contend should be denied protection. I am not in a position to set all of them out here, but they would include such expressive activities as threatening violence, telling deliberate lies, (50) extortion, counselling the commission of a crime and making fraudulent misrepresentations.
a) Protecting constitutionally worthless interests
Nowhere in Irwin Toy does the Court explicitly state that the meaning of freedom of expression to which it commits itself allows for the possibility that expressive content that furthers none of the rationales underlying freedom of expression will be protected. However, it is clear from the analytical framework as a whole that it does. The best evidence of that comes in the Court's treatment of what it calls the "second step" within the framework. It will be recalled that that step entails determining whether or not the government has infringed on freedom of expression either because its purpose was to infringe on it or because that was the effect of the action it took. In describing how this determination is to be made, the Court makes it clear that only those claimants who are obliged to rely on the effect of the government's actions to establish an infringement need show that the expressive activity in question furthers one or more of the rationales for protecting freedom of expression. No such requirement is imposed on plaintiffs who seek to rely on the purpose of the governmental action. It is therefore possible--in fact, it is clearly assumed--that the government will in some cases be found to have violated s 2(b) even though the expressive activity in question furthers none of these rationales.
Additional evidence of this disconnect is provided by Justice McLachlin (as she then was) in her dissenting reasons for judgment in Keegstra. In the course of deciding whether or not hate propaganda should receive the protection of s 2(b), she addressed the argument that hate propaganda should not receive that protection because it does not further any of the rationales for protecting freedom of expression.51 She rejected that argument, not because she believed that hate propaganda does further any of these rationales (although it is clear from the s 1 part of her reasons that she did believe this), but because she refused to accept the premise underlying it--that only expressive activities that further one or more of these rationales qualify for protection.
Justice McLachlin gave several reasons for refusing to accept this premise, and any assessment of the Court's position on this matter has to consider those reasons carefully. Her first objection was that it was inconsistent with the approach to s 2(b) taken in Ford and Irwin Toy. Of the latter case she says, in obvious reference to the analytical framework it prescribed, "[t]he arguments from truth, democracy and self-fulfillment were [there] given a limited role in the interpretation of s 2(b) ... but only in cases where there is no clear government purpose of restricting free expression." (52) She then says that the argument "is essentially circular," by which she means that "[I]f one starts from the premise that the speech covered by s 319(2) [the hate propaganda provision of the Criminal Code] is dangerous and without value, then it is simple to conclude that none of the commonly-offered justifications for protecting freedom of expression are served by it." (53) The third problem she identifies with the argument is "the difficulty of determining when speech has redeeming value." (54) Noting that native leaders may "[make] bitter comments about whites in frustration with governmental failure to recognize land claims," she says that "[e]xperience shows that in actual cases it may be difficult to draw the line between speech which has value to democracy or social issues and speech which does not." (55)
She concludes her reasoning in relation to this issue with the following observations:
Attempts to confine the guarantee of freedom of expression only to content which is judged to possess redeeming value or to accord with the accepted values strike at the very essence of the value of the freedom, reducing the realm of protected dissension to that which is comfortable and compatible with current conceptions. If the guarantee of free expression is to be meaningful, it must protect expression which challenges even the very basic conceptions about our society. A true commitment to freedom of expression demands nothing less. (56)
Do these objections, either collectively or individually, make a convincing case for not limiting s 2(b) protection to those expressive activities that further one or more of the rationales for protecting that right? I think not.57 While the argument from prior authority is clearly accurate insofar as it relies on Irwin Toy, it is based on a highly questionable reading of Ford. In fact, in my view, Ford provides strong support for the view that the rationales should be used to determine the scope of the protection afforded by s 2(b). The Court's decision in that case to accord s 2(b) protection to commercial expression was based very explicitly on the connection between that kind of speech and the individual self-realization rationale. (58)
The complaint about circularity is also unconvincing. There is--or should be, if it is properly explained--nothing in the argument that the meaning of freedom of expression should be functionally linked to the rationales for protecting that right which implies that one begins the analysis by characterizing the expressive activity in question as either dangerous or not dangerous. The analysis begins with the expressive activity itself, and the question is whether or not that activity furthers one or more of the rationales. If it does, it is protected; if it does not, it is not protected. There is nothing circular in that.
It is no doubt true, as Justice McLachlin asserts in her third objection, that it will sometimes be difficult to determine whether a particular expressive activity implicates one of the rationales. But that is a very poor reason for rejecting the argument. For one thing, it is by no means apparent why that determination is any more difficult to make than many others the Charter obliges the courts to make--whether a particular governmental objective is sufficiently important to override a Charter right and freedom, for example, or whether the costs of a particular Charter violation outweigh the benefits. For another, the Court--including Justice McLachlin--has on occasion made precisely this sort of determination an integral part of its s 1 analysis in s 2(b) cases. In Rocket v Royal College of Dental Surgeons of Ontario?5 for example, Justice McLachlin held that promotional advertising by dentists served neither the truth nor the self-government rationale but did serve the self-realization rationale, and, on the basis largely of what she considered to be a relatively significant connection with the latter, struck down the legislation in question. If the Court is able and willing to make such determinations at the s 1 stage, there seems little reason for it to refuse to make them at the definitional stage. And finally, and perhaps most tellingly, the Irwin Toy framework requires precisely this same kind of determination to be made when the claim of infringement is based on the effects rather than the purpose of the impugned governmental action. If the difficulty of making that determination was a good reason not to require judges to make it when defining the reach of s 2(b), the Court would hardly have structured its test for the infringement stage in a way that, at least in some cases, requires judges to make it.
Justice McLachlin's concluding observation about the use of the purposive approach in s 2(b) cases appears to be based on a fundamental misconception about the nature of the approach itself, or about its implications for the scope of protection afforded by s 2(b). Her use of terms like "redeeming value" and "accord with accepted values" suggests that she understands the argument to impose on claimants seeking to have their expressive activities protected the obligation to show that the content of the messages being conveyed is of some beneit to, or is in some sense pleasing to, the majority in society. But that is not what the argument entails. The precise nature of the obligation imposed on claimants would depend on one's understanding of the rationales, of course, but there is nothing in any of the rationales to suggest that they have an inherently majoritarian slant. In fact, if anything, all three of them have an anti-majoritarian slant. The pursuit of truth rationale, for example, is directed very much against the imposition of accepted wisdom on those who prefer other truths. (60) Properly understood, these rationales should work to the benefit of those with dissenting views, not to their detriment, at least insofar as their ability to invoke s 2(b) is concerned. Far from "reducing the realm of protected discussion to that which is comfortable and compatible with current conceptions," (61) as Justice McLachlin fears, using the rationales to generate one's definition of "expression" should provide as strong a guarantee against that happening as there is. (62)
The ability of a rationale-driven approach to defining freedom of expression to give unpopular speech a foothold within s 2(b) is not its only virtue. It also has the merit of being consistent with the purposive approach to Charter interpretation to which the Court first committed itself in 1984 in Hunter v Southam (63) and to which, for the most part, it has remained true since then. (64 ) The whole point of the purposive approach to interpreting rights and freedoms is to ensure that the definitions given to them are faithful to, and hence a function of, the purposes for which they were entrenched in the Charter. In the words of Chief Justice Dickson in Big M Drug Mart, "The meaning of a right or freedom ... [is] to be ascertained by an analysis of the purpose of such a guarantee; it [is] to be understood, in other words, in the light of the interests it was meant to protect." (65) In the case of s 2(b), this means that freedom of expression should be understood in the light of the pursuit of truth, the promotion of self-government and the furtherance of individual self-realization, these being, according to the Court, "the interests [that right] was meant to protect." (66) Activities of an expressive character that can fairly be said to advance one or more of these interests are included within the protected realm, and those that cannot be said to do so are not included.
Whatever one thinks about the merits of the purposive approach, (67) it must surely be conceded that, if the Court identifies what it considers to be the purposes underlying a particular Charter right or freedom, as the Court has done in relation to s 2(b), there is little if anything to commend an interpretation of that right or freedom that results in interests that further none of them receiving constitutional protection. Such interests can fairly be labelled constitutionally worthless. They should not be protected.
b) Otherwise extending the reach of s 2(b) too far
I turn now to a consideration of the approach to defining the reach of freedom of expression that the Court has adopted in place of one based on a faithful application of the purposive approach. According to the Court's approach, protection should be accorded to expressive activities regardless of the content of the message being conveyed. It is on the basis of this approach that such expressive content as threats of violence68 and deliberate lies (69) have been held to be protected by s 2(b). It is also on the basis of that approach that it has been suggested in obiter that committing perjury, defrauding by deliberate misrepresentation, participating in criminal conspiracies and other similar kinds of expressive conduct that Parliament has for a long time prohibited in the Criminal Code should also receive the benefit of such protection.
I have a number of concerns with this approach. (70) The first is that it is not required by the rationale the Court provides for it in Irwin Toy. It will be recalled that the Court wanted to ensure that expressive content that is "unpopular, distasteful or contrary to the mainstream" was protected, "because in a free, pluralistic and democratic society we prize a diversity of ideas and opinions for their inherent value both to the community and to the individual." (71) The Court also invoked the European Court of Human Rights' decision in the Handyside case, in which the connection was made between the "pluralism, tolerance and broadmindedness without which there is no 'democratic society'" and the need to protect "not only ... 'information' or 'ideas' that are favourably received or regarded as inofensive or as a matter of indifference, but also ... those that offend, shock or disturb the State or any sector of the population." (72) Concerns about such expressive content as threats of violence, deliberate lies, perjury and criminal conspiracies can hardly be said to be based on the fact that they are "unpopular, distasteful or contrary to the mainstream," nor on an unwillingness to recognize the value of "a diversity of ideas and opinions" and "pluralism, tolerance and broadmindedness." In fact, it is difficult to see how threats of violence, deliberate lies and extortion come anywhere near being "ideas or opinions" or "information"--or at least the kinds of ideas, opinions and information that the Court in Irwin Toy presumably had in mind--at all.
My second concern is that such an expansive definition of freedom of expression ignores the Court's own admonitions, noted above, that "the Charter was not enacted in a vacuum, and must therefore ... be placed in its proper linguistic, philosophical and historical contexts," (73) and that Charter rights and freedoms should not be interpreted so broadly that they become trivialized. Thirdly, this approach has the potential not only to waste scarce judicial resources by inviting pointless challenges, but also to put governments to the time, trouble and expense of having to make a s 1 justification for no good reason and to encourage courts to further dilute s1's justificatory requirements. (74) Fourthly, if one believes that courts should accord at least some weight to the intentions of the Charter's drafters when engaged in the interpretive process, such an understanding can, I think, safely be said to extend the reach of s 2(b) well beyond what any of the drafters-for that matter, any Canadian-could possibly have envisaged.
Fifthly, that understanding fails to acknowledge that, in addition to identifying the rules by which the people within a particular nation state have chosen to govern themselves, written constitutions have also come to embody, and hence to declare to both the people themselves and the rest of the world, the values that the people consider to be truly fundamental to the way in which they are to be governed. (75) In that sense, written constitutions--and particularly the rights and freedoms that they protect--can be said to perform a very important symbolic function, one of which courts charged with the task of giving effect to them through the process of judicial review should always be mindful. In giving to s 2(b) an interpretation that results in activities such as threatening violence, telling deliberate lies and participating in criminal conspiracies receiving constitutional protection, the Supreme Court has clearly ignored the symbolic role that our Charter plays. While no one can doubt that, in general terms, freedom of expression is a value to which Canadians attach great significance, it is hard to believe that any Canadian would consider the interest in being able to threaten violence, tell deliberate lies and counsel the commission of crimes to be amongst our fundamental values.
Finally, that broad understanding cannot be said to be required by the preference expressed by scholars like Jamie Cameron (76) and several members of the Court--notably Chief Justice Dickson in Keegstra in the course of deciding whether threats of violence deserve the protection of s 2(b) (77) --for leaving the balancing of competing interests in Charter cases to the s 1 stage. That preference is typically justified on the ground that balancing competing interests requires sensitivity to context, and sensitivity to context is more likely to be achieved under s 1 than within the right or freedom itself. (78) I find this line of argument unsatisfactory for two reasons. One is that the Court has shown itself to be both willing and able to balance competing interests during the course of deciding whether or not many of the other rights and freedoms in the Charter have been infringed in particular circumstances. That is especially true of many of the legal rights, including ss 8, (79) 11(a), (80) 11(b),81 11(e) (82) and 12. (83) Of particular significance in this regard is s 7, in which the term "principles of fundamental justice" has effectively been held to import all of the component parts of s 1 as it has come to be understood.84 It has also been evident in the Court's application of "the right to effective representation" understanding of the right to vote in s 3, (85) and was arguably true of the Court's approach to s 15 when Law v Canada (86) held sway as the governing authority. The second reason is based on the rationale given in support of that preference. That rationale assumes that the interest balancing that is going to occur will always require sensitivity to the factual context out of which the Charter claim has arisen. If that assumption is not a valid one in a given case, there is no good reason to have to go to s 1. In my view, sensitivity to context will very rarely if ever be a feature of cases involving such expressive activities as threats of violence, deliberate lies and extortion.
c) Threats of violence and deliberate lies under a purposive approach to s 2(b)
If, as I am suggesting, an understanding of freedom of expression that results in all expressive content being protected should be rejected, and one were to faithfully apply the purposive approach to s 2(b), the question remains as to whether that approach would enable one to justify denying protection to any content-based categories of expression. Would it, for example, justify denying protection to threats of violence and deliberate lies? I believe it would, and that it would do so without having to make any additional refinements to our understanding of the rationales for protecting freedom of expression than those I have proposed above. (87) If our conceptions of democratic self-government, the search for truth and individual self-realization require that these goals be pursued in non-violent ways, then threats of violence have no more place within those conceptions than violent acts, and should also be denied the benefit of s 2(b) protection. Support for this line of reasoning can be found in Justice McLachlin's dissenting reasons for judgment in Keegstra, where she said that threats of violence did not deserve constitutional protection, both because they were "inimical to the rule of law on which all rights and freedoms depend," and because "they undercut ... the role of free expression in enhancing the freedom to choose between ideas (the argument based on truth) or between courses of conduct (the argument based on democracy)." (88)
It might also be a plausible extension of that approach to exclude deliberate falsehoods on that basis. In my view, it is not unreasonable to say to someone seeking the benefit of s 2(b)'s protection that participation in our form of democratic self-government and our search for the truth precludes the use of deliberate falsehoods. Nor is it unreasonable to say to that person that, while we respect and take very seriously the importance of state neutrality on the meaning of the good life, we are prepared to draw the line at advancing one's interests through the deliberate telling of lies.
d) The option of definitional balancing
There is reason to believe, therefore, that a faithful application of the purposive approach to s 2(b) would result in certain content-defined categories of expressive activities falling outside the realm of protected expression. However, even if that were not the case, there is another viable way of denying protection to such categories of expression. That other option, one with which Americans are very familiar, is often called definitional balancing. As many readers will know, the Supreme Court of the United States has denied protection under the free speech clause of that country's Constitution to a range of speech content on the basis of a broad-based or categorical balancing of interests. For example, in New York v Ferber, (89) that court supported its holding that child pornography involving real children fell outside the scope of the free speech clause in part on the basis that, "[it] is not rare that a content-based classification of [unprotected] speech has been accepted because it may be appropriately generalized that within the confines of the given classification, the evil to be restricted so overwhelmingly outweighs the expressive interests, if any, that no process of case-by-case adjudication is required'" (90) While it may be true that definitional balancing has greater appeal in a jurisdiction that lacks a reasonable limits provision like our s 1, the presence of such a provision cannot be said to preclude its use. The fact that, as just noted, balancing competing interests is a common feature of the Court's application of many of the legal rights in the Charter makes that clear.
To readers who are concerned that denying protection to any content-defined categories of expressive activities runs the risk of significantly diluting the salutary effect of the right to freedom of expression in our Charter, I offer this thought: Denying constitutional protection to certain content-defined categories of expression does not have to result in a less vigorous commitment to freedom of expression on the part of the courts. In fact, if American free speech jurisprudence tells us anything, it is that the opposite might be true-that separating out the chaff from the wheat at the first step might well result in more rather than less vigorous protection of free speech interests. There are a number of content-defined categories of speech in addition to child pornography that have been denied protection under the free speech clause of the First Amendment, including deliberate lies, (91) fighting words, (92) obscenity, (93) advertising illegal activities, (94) threats of violence (95) and false and deceptive advertising. (96) Whether or not American courts have been right to deny protection to all of these speech activities, no one can reasonably suggest that denying them protection has prevented American courts from vigorously protecting those speech activities that are protected. (97)
To make explicit what is implicit in the preceding analysis, it is my view that the Court should jettison the first half of the Irwin Toy framework and replace it with an approach that (a) requires the court to be satisfied that the expressive activity at stake is one that furthers at least one of the rationales for protecting freedom of expression; and (b) permits the court to deny protection to that activity if it is satisfied that, to paraphrase the language used by United States Supreme Court in New York v Ferber, the evil the government is seeking to restrict so overwhelmingly outweighs the freedom of expression interests at stake that no process of case-by-case adjudication is required. Such an approach would be more consistent with the purposive approach to interpreting Charter rights and freedoms to which the Court says it is committed, and thereby make the Court's jurisprudence in this area more conceptually coherent. It would eliminate the possibility of providing protection to constitutionally worthless interests. It would ensure that scarce judicial resources are reserved for cases in which such resources should be expended. And it would respect the important symbolic function the Charter plays.
What would the practical consequences of such an approach be? That would depend on both the manner in which the Court defines the reach of the rationales for protecting freedom of expression and the frequency with which it is prepared to find that the standard for exclusion prescribed by New York v Ferber is met. At this point, it is very difficult, if not impossible, to make confident predictions about those matters. That said, given the Court's track record with s 2(b) to this point, I would expect that most of the expressive activities to which the Supreme Court of Canada has afforded constitutional protection, including hate propaganda and pornography, would continue to receive such protection. However, there would be at least some reason to hope, if not expect, that some such activities, including threats of violence and deliberate lies, would not.
I turn now to the second half of the Irwin Toy framework, the half concerned with determining whether or not the impugned governmental action infringes upon a protected s 2(b) interest.
C. Finding an infringement
The Irwin Toy framework is premised on the assumption that there is a conceptual distinction to be drawn between the question of whether a particular interest is protected by freedom of expression and the question of whether the government has conducted itself in such a way as to have infringed on that interest, assuming it to be protected. In answering the first question, the focus is on the it between the interest at stake and the meaning given to freedom of expression-does that meaning catch, and hence protect, that interest? In answering the second, the focus is on the it between the government's conduct and the test for infringing on freedom of expression-is the government's conduct caught by that test?
Different though the two questions are, it is apparent that the test for infringing on freedom of expression that is employed in answering the second will be in part a function of the definition of freedom of expression used in answering the first. It is the test for infringing on that conception of freedom of expression that we are looking for. In this important sense, the two questions are closely related to each other. But they nevertheless remain separate questions, and it is appropriate for them to be treated as such in developing a coherent theory of freedom of expression.
While the Court has constructed its approach to s 2(b) very much on the basis of such a distinction, it has failed to keep the two questions entirely distinct. An integral element of the test devised for an ejects-based infringement is the requirement that the challenger show that the expressive activity subject to government restriction furthers at least one of the rationales for protecting freedom of expression. What this requirement means is that the second step is not concerned solely with the government's conduct; it is also concerned with the it between the expressive activity at issue and the definition of freedom of expression, the gravamen of the first of the two questions.
The failure on the Court's part to keep the two questions distinct is, in my view, a flaw in the Court's approach to finding infringements on freedom of expression, for reasons that I will soon elaborate. But it is not the only flaw in its approach. Another is the equation the Court makes between a content-based and a purpose-based infringement. Related to that problem is the exaggerated importance attached to the distinction between content-based and content-neutral restrictions on expressive activity. And finally, the Court fails to provide a standard by which to measure whether or not the impugned governmental action does in fact restrict an expressive activity protected by s 2(b). Each of these criticisms will be explored in turn.
1. The meaning of a purpose-based infringement
When in Irwin Toy the Court identified both the purpose and the effect of government conduct as potential determinants of infringements on freedom of expression, it was not in any way breaking new ground. It was purporting to apply to freedom of expression what the Court had said earlier in Big M Drug Mart in relation to finding infringements on Charter rights and freedoms. (98) In fact, the Court in Irwin Toy quoted at considerable length from the judgment in that case both to justify looking to both purpose and effect and to provide guidance on the manner in which the infringement inquiry was to be carried out.
Given this heavy reliance on Big M Drug Mart, and in particular the contents of the cited passages, one would have thought that the critical terms "purpose" and "effect" would have been taken to have the same meaning in relation to freedom of expression as they had been given in that judgment. That seems to have been the case with respect to "effect." But it is distinctly not the case with respect to "purpose." It is abundantly clear from Big M Drug Mart that the term "purpose" was being used there in the sense of the objective or goal the government was seeking to achieve when it took the action under attack. In explaining why "both purpose and effect are relevant in determining constitutionality," Chief Justice Dickson says in that case (in one of the passages cited in Irwin Toy),
All legislation is animated by an object the legislature intends to achieve. This object is realized through the impact produced by the operation and application of the legislation. Purpose and effect respectively, in the sense of the legislation s object and its ultimate impact, are clearly linked, if not indivisible. (99)
When Chief Justice Dickson proceeds to examine the purpose of the legislation at issue in that case, the Lord's Day Act, (100) he does so very much in terms of this understanding. It is the objective or goal that the Parliament of Canada was seeking to achieve in enacting the Act that he looks for. That purpose was said to be to compel the observance of Sunday, the Christian Sabbath, as a day of rest by everyone. (101)
This understanding of purpose is also reflected in R v Edwards Books and Art Ltd} (102) another freedom of religion case involving Sunday observance legislation. In that case, which was decided shortly after Big M Drug Mart, the Court concluded that the purpose of the legislation there at issue, Ontario's Retail Business Holidays Act, (103) had been the secular one of providing uniform holidays to retail workers. And it was that purpose, which had initially been attributed to the Act in the context of dealing with a challenge to its validity based in the division of powers in sections 91 and 92 of the Constitution Act, 1867, to which the Court looked when it came to decide whether the Act violated freedom of religion on the basis of its purpose. "The Act," Chief Justice Dickson says, "has a secular purpose which is not offensive to the Charter guarantee of freedom of conscience and religion." (104)
This latter holding in Edwards Books points to another important feature of the Court's approach to purpose-based infringements in the pre-Irwin Toy jurisprudence. Both in Big M Drug Mart and in Edwards Books, the Court distinguishes between "valid" and "legitimate" governmental purposes on the one hand, and "invalid" and "illegitimate" purposes on the other, and makes it clear that it is only the latter that count as purpose-based violations. As Chief Justice Dickson noted in Big M Drug Mart:
Thus, if a law with a valid purpose interferes by its impact with rights or freedoms, a litigant could still argue the effects of the legislation as a means to defeat its applicability or even its validity. In short, the effects test will only be necessary to defeat legislation with a valid purpose; effects can never be relied upon to save legislation with an invalid purpose. (105)
And again, from that case:
In an earlier time, when people believed in the collective responsibility of the community toward some deity, the enforcement of religious conformity may have been a legitimate object of government, but since the Charter it is no longer legitimate. (106)
The critical question, of course, is what makes a purpose "invalid" or "illegitimate." On this, the Court has been less than clear. The closest that Chief Justice Dickson comes in Big M Drug Mart to providing a test or standard is in his assertion that "[t]he assessment by the courts of legislative purpose focuses scrutiny upon the aims and objectives of the legislature and ensures they are consonant with the guarantees enshrined in the Charter." (107) In Edwards Books, as we have seen, he uses the language of offensiveness to the guarantee. A somewhat more explicit articulation of what the Court has in mind comes from Justice Wilson's concurring judgment in Big M Drug Mart when she says, in relation to her s 1 analysis, that "legislation cannot be regarded as embodying legitimate limits within the meaning of s 1 where the legislative purpose is precisely the purpose at which the Charter right is aimed."108 The message appears to be that an illegitimate purpose is one that constitutes an attack upon or denial of the very rationale for protecting either the particular right or freedom in question or Charter rights and freedoms more generally. (109)
If this is an accurate assessment of the Court's approach to purpose-based infringements in the jurisprudence leading up to Irwin Toy, it is not difficult to see the differences between it and the Court's approach to purpose-based infringements in the context of freedom of expression. No longer is "purpose" understood in terms of the government's goal or objective in taking the action under challenge. Instead, it is a measure of whether or not, in pursuing whatever goal the government in question had in mind when it enacted the impugned legislation, the government "has aimed to control attempts to convey a meaning either by directly restricting the content of expression or by restricting a form of expression tied to content." (110) Hence, in deciding in Keegstra that the hate propaganda provision of the Criminal Code infringed s 2(b), Chief Justice Dickson said nothing about the two goals he subsequently held, when he got to the s 1 stage of his analysis, that Parliament was seeking to achieve when it added that offence to the Code111; his reasoning was limited to the fact that "the prohibition in s 319(2) aims directly at words--in this appeal, Mr. Keegstra's teachings--that have as their content and objective the promotion of racial or religious hatred." (112)
Furthermore, we are no longer to speak of "valid" and "invalid" or "legitimate" and "illegitimate" purposes; the operative distinction is now between content-based and content-neutral restrictions. And that distinction calls for a factual rather than normative inquiry: it simply requires the court to determine whether or not the impugned governmental action targets the content of particular messages. Moreover, a finding that a particular restriction is content--based-and hence, according to the new lexicon, purpose-based--is no bar to that restriction being upheld under s 1. In fact, it is arguable on the jurisprudential record to this point, which has seen the Court uphold numerous purpose-based (because tied to content) infringements on s 2(b), that such a finding is irrelevant to the s 1 analysis in s 2(b) cases.
It is tempting to suggest that the Court in Irwin Toy was unaware of the fact that its approach to purpose-based infringements in the context of freedom of expression was very different from that developed in the earlier freedom of religion cases. In form, at least, it purports to be applying the earlier analysis in Big M Drug Mart to freedom of expression. And at no point does it expressly reject the earlier approach in favour of the new and different one. But there are indications in Irwin Toy that the Court knew that the two approaches were different. For example, at one point in the judgment it is said that, "[j]ust as the division of powers jurisprudence of this Court measures the purpose of government action against the ambit of the heads of power established under the Constitution Act, 1867, so too, in cases involving the rights and freedoms guaranteed by the Canadian Charter, the purpose of government action must be measured against the ambit of the relevant guarantee" (113) And then later, "the government cannot have had one purpose as concerns the division of powers, a different purpose as concerns the guaranteed right or freedom, and a different purpose again as concerns reasonable and justified limits to that guarantee. Nevertheless, the same purpose can be assessed from different standpoints when interpreting the division of powers, limitations of a guarantee, or reasonable limits to that guarantee." (114)
Whether or not the majority intended to fashion a new and different approach to purpose-based infringements for freedom of expression cases, the question must be asked--is this new approach preferable to the one articulated in the earlier freedom of religion cases? I think not, and for a number of reasons. There is, first, the rather obvious point that, in spite of its claim to the contrary, the Court in Irwin Toy is not merely assessing the same purpose from different standpoints when it defines purpose one way at the point at which it decides whether freedom of expression has been infringed and another way at the point at which it decides whether that infringement can be justified. It is speaking of and using two completely different meanings of the word. Secondly, and more importantly, it is misleading to use the word "purpose" to describe a feature of the impugned governmental action--whether or not it restricts particular content-that relates far more to means than to ends. In Irwin Toy, the government's objective--as the majority recognized when it got to s 1--was to protect children and their families from being manipulated into buying children's toys and other products; the restrictions imposed on the advertising of such products--defined in terms of the content of such advertising--constituted the means to that end.
There is, finally, a good deal to be said for an approach to purpose-based infringements of Charter rights and freedoms that incorporates the notion of an invalid or illegitimate purpose, at least if that notion is understood in the narrow sense in which it appears to have been used in the freedom of religion cases. It will be, or at least should be, a very rare case nowadays in which a court finds that a government has been guilty of a direct attack upon or denial of the very rationale for protecting a Charter right. That is so not only because governments in this day and age should know better than openly to act out of what amounts to a mala fides desire to deprive people of their constitutional rights, but also because courts are always going to be reluctant to attribute bad faith to governmental bodies, and especially to elected bodies. (115) However, as Chief Justice Dickson argued in Big M Drug Mart, (116) the courts should be in a position to make such a finding, and to hold that the governmental action in question is unconstitutional regardless of the benefits that might be said to flow from it and regardless of the particular right or freedom at stake. No government should be entitled to act for a purpose that is wholly at odds with the purpose for which a Charter right or freedom, or the Charter as a whole, was entrenched. The approach to purpose-based infringements adopted in Big M Drug Mart incorporates that principle; its counterpart in the context of freedom of expression does not, and to that extent the latter approach is flawed.
It must also be said that it makes little sense to have different approaches to this question being adopted in relation to two rights, freedom of religion and freedom of expression, that are so closely linked together. This linkage is not only a product of the fact that freedom of religion and freedom of expression are found in the same section of the Charter and share the same "fundamental freedom" label. It is a function also of the shared historical origins of these two rights--freedom of expression can readily be understood as having originated as a political principle in the religious persecution of heretics and blasphemers--and of the shared philosophical premises underlying the two rights--particularly the commitment to individual self-realization. (117) And it arises from the obvious but often forgotten fact that where religious expression is involved, claims can be made simultaneously under both provisions. On what ground can it plausibly be argued that the two claims should be analyzed in two very different ways, with the government's purpose changing from claim to claim as a function of the right being invoked? All things considered, there seems little to commend the Court's interpretation of purpose in the context of freedom of expression. The sooner it is altered to coincide with that articulated in Big M Drug Mart and Edwards Books, the better.
2. The special rule applicable to effects-based infringements
I turn now to another of the laws in the Court's approach to finding infringements on freedom of expression, the fact that an integral part of that approach, insofar as effects-based infringements are concerned, is an inquiry into the connection between the expressive activity in question and the rationales for protecting freedom of expression. It will be recalled that challengers seeking to establish an effects-based infringement--but not a purpose-based infringement--must show that their particular expressive activity furthers one or more of those rationales.
My complaint is not that challengers should be obliged to show that the expressive activities for which they seek protection under s 2(b) further one or more of the rationales for protecting freedom of expression. I have argued above that they should be so obliged. What is troubling about the Irwin Toy framework is that the obligation to make this connection lows from the nature of the infringement a particular challenger is seeking to establish. If the obligation is to be imposed--as I have argued above it should be--its imposition should not depend on whether the infringement is purpose-based or effects-based; it should be imposed on all challengers regardless of the nature of the infringement. Moreover, it should be imposed at the definitional stage, not as part of the infringement inquiry.
This latter point may be one of form more than substance, but important nevertheless if one is concerned, as the Court should be, with the conceptual integrity of its analytical processes. As noted earlier, there is a clear conceptual difference between the question of whether a particular expressive activity is entitled to the protection of s 2(b) and the question of whether, if it is, the impugned governmental conduct infringes on it. The question of whether an expressive activity furthers one of the rationales for protecting freedom of expression is, or should be, a question of the first kind, not the second. It relates to the scope of protection afforded by--or simply, the definition of--the freedom of expression granted by s 2(b). It belongs at the definitional stage.
Had this question been incorporated in the definitional stage, it is highly unlikely that the Court's analytical framework would have distinguished in the ways it does between purpose-based and effects-based infringements. But it was not, and that distinction constitutes an integral part of the infringement inquiry. Why is that distinction a problem? (118) It is a problem first because it allows for the possibility that governments will be found guilty of infringing s 2(b) in circumstances in which the activities of the challengers are in no way linked to any of the rationales for protecting freedom of expression. As I have argued above, that such a possibility exists is, or should be, troubling from the standpoint of both theory and practice. What theoretical justification can there be for according constitutional protection to activities that bear no relation to the reasons for entrenching the right in question in the Constitution? Surely none. And as a practical matter, why should governments be required to defend their actions in the courts in such circumstances? Again, there seems no reason for it.
This distinction is also problematic because of another possibility for which it allows--the possibility that exactly the same expressive activity will be protected by s 2(b) in one case and not protected in another. Assume that A and B both wish to engage in the same expressive activity. Assume further that this expressive activity satisfies the Irwin Toy definition of "expression"--that it amounts, in other words, to non-violent activity that conveys or attempts to convey a meaning--but does not further any of the rationales for protecting freedom of expression. Now assume that both A and B are prohibited by government from engaging in this activity, A by a law tied in some sense to content, B by a law not so tied. Applying the Irwin Toy approach to infringements on freedom of expression, this same expressive activity would receive the protection of s 2(b) when A engaged in it but not when B did. That can hardly be said to make sense.
The distinction between purpose--based and effects--based infringements is further weakened by the absence of any attempt on the Court's part to justify it. Irwin Toy itself is completely silent on the matter of justification--the distinction is simply announced. (119) And one looks in vain for any justification in the many subsequent articulations of this part of the Irwin Toy framework. The closest any member of the Court has come to providing one that I have found is a comment, made very much in passing, by Justice McLachlin in Committee for the Commonwealth of Canada v Canada (120) a case concerned with access to public property, to the effect that the obligation to link the expressive activity with the rationales imposed on those seeking to establish an effects--based infringement makes sense in light of the expansive definition of "expression" that the Court has adopted. Such a comment hardly provides a compelling justification, however, because it fails to respond to the obvious question of why the obligation should be limited to effects-based infringements. (121)
It should be noted that the use of this distinction in freedom of expression cases serves to compound even further the difference between the Court's approach to finding infringements in such cases and the approach it uses in freedom of religion cases. No such distinction is drawn, or has even been hinted at, in the Court's approach to the latter. There, the question of definition, of scope of coverage, is kept separate from the question of infringement, and the same activities are protected whether the infringement is purpose--based or effects-based. Hence, there is no suggestion in Edwards Books and Art, where the infringement was of the latter kind, that the challengers had any additional obligations in respect of coverage to those imposed on their counterparts in Big M Drug Mart, where the infringement was purpose--based. The question of coverage had already been settled.
3. The distinction between content--based and content-neutral infringements
Integral to the Court's approach to finding infringements in freedom of expression cases is the underlying distinction between content-based and content-neutral restrictions on expressive activities. As we have seen, a content-based restriction has been equated to a purpose-based infringement and, as a corollary, a content-neutral restriction to an effects-based infringement. As other commentators have pointed out, the origins of this distinction, and of its use by the Court as an analytical device, lie in American First Amendment law and theory. (122) It is no surprise, therefore, that the Court in Irwin Toy invokes in support of its use in Canada American theorists such as Archibald Cox and Thomas Scanlon, for both of whom the distinction was of profound importance. (123)
Important as that distinction has been to American thinking about freedom of speech, two points need to be made about its use in the approach that American courts take to free speech cases. The first is that that distinction only becomes important at the stage of deciding whether or not a particular limit on free speech can be justified; put simply, governments will be put to a much higher standard when they seek to justify a content-based limit than when they seek to justify a content-neutral limit. That distinction plays no role in deciding whether or not free speech has been limited. (124) By incorporating the distinction into its analysis of whether or not freedom of expression has been limited, the majority in Irwin Toy has therefore misapplied it. (125)
The second point is that the distinction is by no means universally accepted in the United States as a useful analytical device, and has in fact been subjected to trenchant criticism by leading First Amendment scholar Steven Shiffrin in his book The First Amendment, Democracy and Romance. (126) The nature of that criticism is such that it is equally relevant to the use of the distinction in Canada as to its use in the United States.
Shiffrin's attack on the content-based versus content-neutral distinction is part of a general assault on the notion that free-speech problems can and should be resolved by the application of a small number of rules borne of an even smaller number of distinctions. His thesis-one that resonates generally with the approach Canadian courts have taken to freedom of expression-is that free speech problems come in so many different shapes and sizes that any such rule-fixated approach to their resolution is doomed to failure. Shiffrin does not reject the use of rules completely; he rejects the use of nothing but rules. This thesis is based primarily on his conviction that "some pockets of first amendment law involve so many competing variables that any rule would lead to intolerable results in too many concrete cases," (127) but also on his belief that rules can be less protective of free speech values than ad hoc balancing. Finally, he contends that, because the application of many rules necessarily entails some degree of ad hoc balancing, the line between the two is not as sharp as people assume.
Shiffrin devotes considerable time to the distinction between content-based and content-neutral restrictions on expression because, as he is quick to acknowledge, that distinction has played, and continues to play, such an important role in American thinking about freedom of speech. It formed the basis of much of the writing of the influential First Amendment scholar, Melville Nimmer, (128) and lies at the heart of Laurence Tribe's assertion in his textbook on American constitutional law that there are "two distinct approaches to the resolution of first amendment claims; the two correspond to the two ways in which government may 'abridge' speech." (129) Other scholars too, including, as the Supreme Court noted in Irwin Toy, Archibald Cox (130) and Thomas Scanlon, (131) as well as John Hart Ely, (132) have invoked the distinction in their writings.
The gist of Shiffrin's complaint about the distinction is captured in his opening assertion about it: "This distinction," he says, "is a proxy for something important in first amendment law, but the matter for which the proxy has been created is only one of many important aspects of first amendment law, and the proxy is quite imperfect." (133) The "something important" for which the distinction is a proxy, according to Shiffrin, is that "government may not abridge speech merely because it is biased against that speech." (134) The distinction is believed to get at that "something important" by separating those restrictions on speech that are likely to low from such a bias from those that are not. That belief is grounded in the assumption that governments are more likely to be acting out of bias when they explicitly restrict content than when they do not.
Shifrin questions the validity of this assumption. In his view, content-based restrictions on expression may be enacted for reasons--good reasons--having nothing whatsoever to do with a simple government bias against speech; he cites as examples the criminalization of perjury, blackmail and fraud. By the same token, he says, government bias can play an important role in the enactment of content-neutral restrictions. For example, a ban on using loudspeakers or anonymous pamphleteering in particular locales could easily be motivated by a desire to stile dissentient political speech.
More importantly, argues Shiffrin, the prohibition against bias--or, as he and others often describe it, point-of-view discrimination--is by no means absolute. In the running of public schools, public libraries, public concert halls, and public funding agencies, government inevitably prefers some ideas or points of view to others--what courses to teach, what books to purchase, and so on--and, by and large, such preferences are not taken to violate the First Amendment. As Shiffrin puts it, "[g]overnment plays an enormous role in the intellectual marketplace. By subsidizing some speech and shunning other speech, it rejects the equal status of ideas." (135) Nor, he claims, is society's acceptance of these deviations from the principle of content-neutrality limited to contexts in which the government is operating such institutions; it extends to at least some contexts in which speech is being suppressed through regulation or prohibition. He invokes the American courts' approach to obscenity and other forms of unprotected speech as support for this assertion. He also notes the willingness of American judges to assign different degrees of worth or value to different kinds of speech; commercial speech and profane speech are his examples here. (136)
In short, says Shiffrin, "[a]s a proxy for finding government hostility to points of view, each half of the distinction is under-inclusive and over-inclusive. Moreover, First Amendment law is not exclusively, or even mainly, concerned with government hostility to particular points of view. Finally, government hostility to particular points of view is often acceptable, and evaluations of the value of speech run through the fabric of first amendment law." (137) Hence, he concludes, the distinction between content-based and content--neutral restrictions should not be--and, as he goes on to establish by careful examination of selected areas of American free speech jurisprudence, is in fact not--as important to American thinking about free speech problems as accepted wisdom would have us believe it is.
Shriffin is also of the view that the line between content-based and content-neutral restrictions on free speech is not as bright as it is assumed to be. Using the aforementioned case of New York v Ferber to help him make this point, he notes that, although the prohibition against distributing films showing children engaged in sexual activity there at issue was in one sense content-based, in another it was not. (138) It was content-based because the prohibited speech was defined in terms of content. It was not content-based because, as generally understood, the government's reason for legislating was not based on a concern about the message being conveyed--it wanted to remove the incentive for the sexual exploitation of children that is part and parcel of the making of such films. And the distinction remains a tricky one to make, he says, when another possible rationale for the legislation--avoiding the psychological harm to the children involved that comes from knowing that the film exists and might be viewed by others--is taken into account. Again, there is an obvious connection to content, but the concern is again unrelated to the impact of the message being conveyed on those who view the films. (139)
The importance of these criticisms of the distinction, if one accepts them, is ultimately a function of the significance of the role which the distinction plays in the resolution of free speech problems. The less significant that role is, the less important the criticisms are. In the United States, that role is generally thought to be a very significant one. As noted above, governments are required to satisfy a much more stringent test if they seek to justify a content-based limit on free speech than if they seek to justify a content-neutral limit.
On the face of it, the role of the distinction in Canada, while different, also looks to be significant. According to the approach to finding infringements on s 2(b) prescribed by Irwin Toy, challengers who can show the government to have been guilty of a content-based restriction on freedom of expression can force the government to a s 1 defence of that restriction without having to establish any connection between the expressive activity in question and the rationales for protecting freedom of expression, whereas challengers who are forced to rely on the showing of a content-neutral restriction do have to establish such a connection. Because the Court has seldom had to concern itself with the latter form of restriction, having shown itself to be more than willing to accept that restrictions are content-based, we do not yet know whether this difference will affect the outcome of many s 2(b) cases; much will presumably depend on the interpretation eventually given to the rationales themselves. But the potential is certainly there for the difference in approach to have a bearing on the outcome of at least some cases. And to that extent, Shiffrin's criticisms must be borne in mind.
4. What amounts to a restriction on freedom of expression?
One of the more curious features of the second part of the Irwin Toy framework is that it provides no guidance on how courts should go about deciding whether or not the impugned governmental action actually restricts freedom of expression. (140) The framework simply assumes that it does, and addresses instead the question of what kind of restriction has resulted from that action. That would not be a problem if it were always and obviously the case that the impugned action in challenges based on s 2(b) does restrict freedom of expression. However, as American experience demonstrates all too clearly, that is not always and obviously the case. The free-speech jurisprudence of that country includes a range of cases in which the United States Supreme Court has had to grapple with the first order question of whether or not any infringement of freedom of speech has in fact occurred. That question has arisen in the context of challenges to legislation which disentitles people from being remunerated for engaging in certain kinds of expressive activities, (141) legislation that conditions entitlement to a government benefit on foregoing certain kinds of expressive activities, (142) and legislation that applies pressure of one kind or another on people to refrain from engaging in certain kinds of expressive activities. (143) And the results in those cases have not been uniform: in some of them, that court has held that the legislation did infringe on freedom of speech, while in others, it has held that the legislation did not.
The failure of the Court to address this question in Irwin Toy is surprising, because it had already addressed precisely that same question in the context of freedom of religion. Of particular importance in that regard are the aforementioned decisions in Big M Drug Mart and Edwards Books & Art. In the former, Chief Justice Dickson held that the essence of "freedom" was "the absence of coercion or constraint," (144) and went on to define coercion as including "not only such blatant forms of compulsion as direct commands to act or refrain from acting on pain of sanction," but as also encompassing "indirect forms of control which determine or limit alternative courses of action available to others." (145) In the latter, he held that "[i]t matters not ... whether a coercive burden is direct or indirect, intentional or unintentional, foreseeable or unforeseeable. All coercive burdens on the exercise of religious beliefs are potentially within the ambit of s 2(a)." (146) He then added, however, that not all such burdens are "offensive to the constitutional guarantee of freedom of religion." (147) "For a state-imposed cost or burden to be proscribed by s 2(a)," he said, "it must be capable of interfering with religious belief or practice. In short, legislative or administrative action which increases the cost of practising or otherwise manifesting religious beliefs is not prohibited if the burden is trivial or insubstantial." (148)
If those principles are thought to be appropriate in the context of freedom of religion, there seems little reason not to incorporate them into the Court's approach to freedom of expression. The two sit side by side within the same category of rights and freedoms in the Charter. The same notion of freedom that is said by Chief Justice Dickson to underlie the former--"the absence of coercion or constraint"--can be said to underlie the latter. They share similar historical and philosophical roots. And, at least in the sphere of religious expression, they protect the same interests.
It should be noted that incorporating these or a similar set of principles into the existing Irwin Toy framework would not be difficult. All the Court would have to do is add to the second part of the framework a preliminary step in which the question is asked whether the impugned governmental action does or does not restrict, or infringe upon, the protected expressive activity at issue, and stipulate that that question is to be answered on the basis of those principles. If that question is answered in the negative, the challenge ends there; if the question is answered in the affirmative, the court can move on to the rest of the analysis called for by the framework.
That said, it is my hope that the Court will jettison the second part of the existing framework, and replace it with the following simplified analysis: (1) Does the impugned governmental action restrict freedom of expression (relying on the principles developed in the context of freedom of religion)? If it does not, the challenge would be dismissed; if it does, the court proceeds to the next step. (2) Is the purpose underlying the impugned governmental action illegitimate or invalid (in the sense in which those terms are used in the context of freedom of religion)? If it is, the impugned governmental action should be struck down; if it is not, the court proceeds to s 1.
I should perhaps add that the addition of this step would not have changed the outcome of any of the s 2(b) cases that the Supreme Court of Canada has decided to this point under the Irwin Toy framework. In fact, it would only be if and when the Court is called upon to resolve the kinds of issues noted above that the U.S. Supreme Court has had to grapple with that this step would play a meaningful role in determining the outcome. From a strictly practical standpoint, therefore, its significance is likely to be minimal. However, if one is concerned about the conceptual integrity of the Court's approach to Charter adjudication in general and to freedom of expression in particular, as I believe we all should be, incorporating it in the analysis has very real and significant benefits.
In summary, the Irwin Toy framework is flawed for a number of reasons. While the meaning it gives to freedom of expression appropriately excludes violent expressive acts, the rationale that has been given for excluding such acts is far from compelling. That meaning is otherwise overly broad in its reach, is not required by the reasons given in support of it, is inconsistent with the purposive approach to defining rights and freedoms, provides constitutional protection to constitutionally worthless interests, encourages needless litigation, and fails to take into account the important symbolic function the Charter plays. The analytical steps the framework prescribes for the determination of whether freedom of expression has been infringed incorporates an ill-advised change in the understanding of what constitutes a purpose-based infringement, imposes for no good reason different obligations on claimants depending on whether they are challenging an effects-based or purpose-based limit on their freedom of expression, is based on a distinction of questionable analytical utility and fails to incorporate a critically important analytical step.
I would like to close with a word or two about the likelihood of the Court changing its position in relation to any of these problematic features of the Irwin Toy framework. There certainly have not been any signals from the Court in any of its recent cases that it is prepared to engage in a major rethink of any of them. On the contrary, as its recent decision in the Criminal Lawyers' Association (149) case demonstrates, the Court is continuing to call that framework into use quite happily.
That said, one can find the odd signal that individual members of the Court might be open to a rethink, if not of all of the framework's features, then at least of some of them. In his concurring reasons for judgment in a recent freedom of religion case, Multani v Commission scolaire Marguerite-Bourgeoys, (150) Justice LeBel said that the Court "[has] never definitively established that [relying on s 1 of the Charter] is the only way to reconcile competing or conflicting fundamental rights," and that as a matter of constitutional policy, he thought that a more flexible approach to balancing competing rights, one that allowed for more such balancing to take place within the rights themselves, made sense, at least in some contexts. Those sentiments suggest that he at least might be open to denying certain content-based categories of expression the protection of s 2(b) on the basis of what I referred to above as definitional balancing.
A willingness to incorporate a balancing of interests into the definition of s 2(b) itself is evident in the Criminal Lawyers' Association case. That case was not what I have called the standard or typical s 2(b) case, one in which the claim is made that a government has restricted the ability of people to express themselves in the manner they wished to. It was not, therefore, the kind of case for which the Irwin Toy framework was originally designed. (151) Nevertheless, the case did require the Court to consider, as the Court itself put it, "essentially a question of how far s 2(b) protection extends." (152) The specific issue it raised was the extent to which s 2(b) can be said to guarantee access to government-held documents. In reasons for judgment co-authored by Chief Justice McLachlin and Justice Abella, the Court resolved that issue by holding that s 2(b) does protect access to some such documents, but "only where it is shown to be a necessary precondition to meaningful expression, does not encroach on protected privileges, and is compatible with the function of the [government] institution concerned." (153) It is clear that the imposition of the latter two of these three conditions reflects a decision on the Court's part to give greater weight to the interests they embody--the protection of privileged communications and the proper functioning of governmental institutions-than the interest in being able to participate in "meaningful expression." And given that those conditions are imposed as part of the process of defining the reach of the protection that s 2(b) affords, their imposition provides clear evidence of the Court's engaging in definitional balancing. If the Court is willing to engage in that process in the context of access to government-held documents, perhaps it will be willing to do so in other contexts as well.
The Criminal Lawyers' Association case is important for a second reason. It is apparent that the first of the three conditions that form part of the Court's delineation of the claims of access to government-held documents that are protected by s 2(b)--the need to show that such access is "a necessary precondition to meaningful expression"--reflects a belief on the Court's part that there is a need to connect the expressive interest for which constitutional protection is sought to the rationales underlying freedom of expression. In fact, the Court is quite explicit about this: it says, "[t]o show that access would further the purposes of s 2(b), the claimant must establish that access is necessary for the meaningful exercise of free expression on matters of public or political interest." (154) Again, one is tempted to ask--if that is an appropriate requirement in this context--why not more generally?
Finally, there is an indication in one of the Court's recent judgments that its commitment to protecting all expressive activities regardless of the content of the message being conveyed might be waning somewhat. In her reasons for judgment in the Greater Vancouver Transit Authority case, in which she was speaking for seven members of the Court, Justice Deschamps said the following in her introduction to her s 2(b) analysis,
s 2(b) of the Charter is not without limits and governments will not be required to justify every restriction on expression under s 1.... The method or location of expression may exclude it from protection: for example, violent expression or threats of violence fall outside the scope of the s 2(b) guarantee, and individuals do not have a constitutional right to express themselves on all government property. (155)
The significance of this passage lies in its inclusion of threats of violence in the list of examples of expressive activities that are excluded from the reach of s 2(b). No mention is made by Justice Deschamps of the fact that by including them in the list, she is both overruling the specific holding of the majority of the Court in Keegstra and departing from the Court's until then apparently firm general conviction that no expressive activity should be denied s 2(b)'s protection because of the content of the message being conveyed. Nor does she provide any rationale for excluding them. Perhaps their inclusion in the list was an oversight. However, if it wasn't, then the possibility presumably exists that other content-defined expressive activities will also eventually come to be excluded, and the meaning of freedom of expression will end up being somewhat narrower than it now is.
All of which is to say, I suppose, that there is reason still to hope that the Court might one day get the building blocks of its approach to freedom of expression right.
(1)  1 SCR 927, 58 DLR (4th) 577 [Irwin Toy cited to SCR]. The case involved a challenge to Quebec legislation imposing severe restrictions on the ability of businesses in that province to advertise their products to young children. The Court held that the legislation restricted freedom of expression, but upheld it under s 1 of the Charter as a "reasonable limit" on that freedom. I should explain here that I will be using the term "the Court" in relation to the Supreme Court's judgment in this case even though the analytical framework that emerged from it was the product of the work of only three of the five members of the Court that participated in its resolution (Chief Justice Dickson and Justices Lamer and Wilson). The other two judges who participated in its resolution (Justices McIntyre and Beetz) did not comment on the framework's suitability (and dissented in the result). I have used this terminology in part for the sake of convenience, but also because the framework was very quickly accepted by all of the other members of the Court.
(2) Canadian Charter of Rights and Freedoms, Part 1 of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [the Charter].
(3) See e.g. J Cameron, "The Original Conception of Section 1 and Its Demise: A Comment on Irwin Toy Ltd. v Attorney-General of Quebec" (1989), 35 McGill LJ 253; D Gibson, "Constitutional Law-Freedom of Commercial Expression under the Charter-Legislative Jurisdiction over Advertising-A Representative Ruling-Irwin Toy Ltd. v Attorney-General of Quebec" (1990) 69 Can Bar Rev 339; T Macklem, "Toying with Expression: Irwin Toy Ltd. v Quebec (Attorney-General)" (1990) 1 Sup Ct L Rev (2d) 547; L Weinrib, "Does Money Talk? Commercial Expression in the Canadian Constitutional Context" in D Schneiderman, ed, Freedom of Expression and the Charter (Toronto: Carswell, 1991) 341; D Lepofsky, "The Supreme Court's Approach to Freedom of Expression: Irwin Toy Ltd. v Quebec (Attorney General) and the Illusion of Section 2(b) Liberalism" (1993) 3 NJCL 37; and R Moon, "The Supreme Court of Canada on the Structure of Freedom of Expression Adjudication" (1995) 45 UTLJ 419. I should acknowledge that not everyone who has written about the approach taken to freedom of expression in Irwin Toy has been critical of it. For a sampling of some of the positive assessments, see J Ross, "The Protection of Freedom of Expression by the Supreme Court of Canada" (2003) 19 Sup Ct L Rev (2d) 81; S Anand, "Beyond Keegstra: The Constitutionality of the Willful Promotion of Hatred Revisited" (1998) 9 NJCL 117; and M Tilliard, "Commercial Expression Comes of Age: The Path to Constitutional Recognition Under the Charter of Rights" (1990) 28 Alb L Rev 604.
(4) I have been unable to find a single case in which the Supreme Court of Canada has even mentioned, let alone addressed, any of the criticisms levelled against the framework by those early commentators.
(5) 2010 SCC 23,  1 SCR 815 [Criminal Lawyers' Association].
(6) I should make it clear that, in assessing the soundness of these understandings, I am concerned not only with the understandings themselves-that is, the conclusions that the Court arrives at in relation to each of the two elements of its approach to s 2(b)-but also with the reasoning the Court employs in arriving at those conclusions. A truly sound understanding is therefore one that not only can be persuasively defended, but that the Court itself has persuasively defended.
(7) Irwin Toy, supra note 1 at 978.
(8) As will become apparent, this assertion is somewhat misleading because of the manner in which the Court requires effects-based restrictions of freedom of expression to be dealt with (see infra notes 114-117 and accompanying text). However, it is the case that the meaning of freedom of expression is primarily provided by these two sentences.
(9) See e.g. Ford v Quebec (AG),  2 SCR 712, 54 DLR (4th) 577 [Ford] and R v Keegstra,  3 SCR 697,  2 WWR 1 [Keegstra cited to SCR].
(10) For the purposes of this paper, I am content to assume both that these are all appropriate rationales to ascribe to freedom of expression as a constitutionally protected right in the Charter, and that there are no others. I acknowledge that the first of these propositions (at least as applied to the free speech clause in the First Amendment, US Const amend I) has been called into question by some American scholars (see e.g. A Meiklejohn, Free Speech and Its Relation to Self-Government (New York: Harper, 1948) and M Redish, "The Value of Free Speech" (1982) 130 U Pa L Rev 591). I also acknowledge that the second proposition has been called into question both here in Canada (see e.g. Gibson, supra note 3 at 344; Lepofsky, supra note 3 at 73-74; and K Dubick, "The Theoretical Foundation for Protecting Freedom of Expression" (2001) 13 NJCL 1) and in the United States (see e.g. F Schauer, Free Speech: A Philosophical Enquiry [Cambridge: Cambridge University Press, 1982], ch 1). It is unnecessary for me to address the merits of these propositions here because neither the number nor the identity of the rationales for protecting freedom of expression affects the main thrust of the critique of the Irwin Toy framework that I advance in this paper. However, I will say that I agree strongly with Richard Moon's contention that the Supreme Court of Canada has a great deal more work to do in fleshing out what it understands the nature and scope of the rationales underlying s 2(b) to be (see R Moon, "Drawing the Lines in a Culture of Prejudice: The Restriction of Hate Propaganda" (1992) 26 UBC L Rev 99). That is particularly true, in my view, of the individual self-realization rationale, which as currently understood by the Court appears to have no limits.
(11) The Court has attempted to fully merge its approach to this kind of s 2(b) case into the Irwin Toy framework in its more recent jurisprudence (see Montreal (City of) v 2952-1366 Quebec Inc, 2005 SCC 62,  3 SCR 141 and Criminal Lawyers' Association, supra note 5), but that framework played a limited role in the early cases in this area (see Committee for the Commonwealth of Canada v Canada,  1 SCR 139, 77 DLR (4th) 385 and Ramsden v Peterborough (City of),  2 SCR 1084, 106 DLR (4th) 233).
(12) The leading case in this area (which refers to all of the earlier cases) is Baier v Alberta, 2007 SCC 31,  2 SCR 673.
(13) For examples of cases in which the Court has been called upon to deal with this kind of issue, see Lavigne v Ontario Public Service Employees Union,  2 SCR 211, 81 DLR (4th) 545; RJR MacDonald Inc v Canada (AG),  3 SCR 199, 127 DLR (4th) 1; and Canada (AG) v JTI Macdonald Corp, 2007 SCC 31,  2 SCR 610.
(14) Irwin Toy, supra note 1.
(15) Edmonton Journal v Alberta (AG),  2 SCR 1326, 103 AR 321 [Edmonton Journal].
(16) Reference Re ss 193 and 195.1(1)(c) of the Criminal Code (Man),  1 SCR 1123,  4 WWR 481 [Prostitution Reference cited to SCR].
(17) Keegstra, supra note 9.
(19) R v Butler,  1 SCR 452, 89 DLR (4th) 449 [Butler cited to SCR].
(20) R v Zundel,  2 SCR 731,95 DLR (4th) 202.
(21) Haig v Canada; Haig v Canada (Chief Electoral Officer),  2 SCR 995, 105 DLR (4th) 577.
(22) R v Sharpe, 2001 SCC 2,  1 SCR 45.
(23) Baier v Alberta, supra note 12.
(24) Prostitution Reference, supra note 16 at 1182-83, Lamer J (as he then was).
(25) Supra note 1 at 968.
(27) Palko v State of Connecticut, 302 US 319, 58 S Ct 149 at 327 (1937), cited ibid at 968-69.
(28) Ibid at 969.
(29) The passage is from Handyside v United Kingdom,  ECHR 5493/72, 1 EHRR 737 (1976) [Handyside], cited ibid at 23.
(30) Ibid at 969-70.
(33) RWDSU v Dolphin Delivery Ltd,  2 SCR 573 at 588, 33 DLR (4th) 174.
(34) Irwin Toy, supra note 1 at 970.
(35) David Lepofsky makes the same criticism in his comment on Irwin Toy, supra note 3 at 51.
(36) Keegstra, supra note 9 at 732.
(37) Montreal (City of) v 2952-1366 Quebec Inc, 2005 SCC 62,  3 SCR 141 at para 72.
(38) I am grateful to my research assistant Brandon Langhjelm for pointing this out to me.
(40) Ford, supra note 9.
(41) R Moon, The Constitutional Protection of Freedom of Expression (Toronto: University of Toronto Press, 2000) at 24.
(42) Moon, supra note 3 at 429.
(43)  1 SCR 295 at 344, 18 DLR (4th) 321 [Big M Drug Mart cited to SCR].
(44) Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, reprinted in RSC 1985, App II, No 5.
(45) Cameron, supra note 3 at 268.
(46) See e.g. s 4(2) (democratic rights), s 6(3) and (4) (mobility rights), s 15(2) (equality rights, at least as now understood), s 20(1)(b) (official languages), and s 23(3) (minority language educational rights).
(47) Gibson, supra note 3 at 342.
(49) L Tribe, American Constitutional Law (Mineola, NY: Foundation Press, 1978), ch 12-17.
(50) I should make it clear that when I use the term "deliberate lies" (or "deliberate falsehoods") in this article, I mean deliberate misstatements of fact. I do not mean the expression of misguided opinions or ideas, even for nefarious purposes. The latter, in my view, should not be denied protection under s 2(b).
(51) Supra note 9. It should be noted that this argument was not addressed by Chief Justice Dickson in his reasons for judgment.
(52) Ibid at 840-41.
(53) Ibid at 841.
(55) Ibid at 842.
(57) I should note that both Jamie Cameron and Dale Gibson agree with the Court's position on this issue (see Cameron, supra note 3 at 255 and Gibson, supra note 3 at 343).
(58) Supra note 9 at 766-67.
(59)  2 SCR 232, 71 DLR (4th) 68.
(60) In the words of Justice Holmes in his dissenting judgment in Abrams v United States, 250 US 616, 40 S Ct 17, at 630 (1919), upon which the pursuit of truth rationale is largely based, "when men have realized that time has upset many lighting faiths, they may come to believe even more than they believe the very foundations of their own conduct that the ultimate good desired is better reached by free trade in ideas-that the best test of the truth is the power of the thought to get itself accepted in the competition of the market, and that truth is the only ground upon which their wishes can be safely carried out." It is clear from the context in which he used these words that the "men" to which Justice Holmes referred were those holding the reins of power at a given moment in time, and that the "ideas" with which their majoritarian "truth" would compete in the market were non-majoritarian in origin.
(61) Keegstra, supra note 9 at 842.
(62) It may be true, as one of the external reviewers suggested, that the Supreme Court of Canada's track record in using these rationales within the context of s 1 of the Charter (as part of the highly contextualized analysis the Court insists courts must engage in under that provision) has been less than inspiring. However, the remedy to that problem is not, in my view, to have the courts abandon all references to the rationales for protecting freedom of expression when they are called upon to adjudicate cases under that provision. It is to urge them to take those rationales, and in particular the anti-majoritarian potential they embody, more seriously.
(63) Hunter et al v Southam Inc,  2 SCR 145, 11 DLR (4th) 641.
(64) I say " for the most part" because there are some Charter rights and freedoms-s 7 would be a good example-in respect of the interpretation of which it is difficult to find clear evidence of the purposive approach.
(65) Supra note 43 at 344.
(67) It should be noted that Frederick Schauer, one of the leading American scholars on the meaning and scope of free speech as a constitutional right, and someone on whom members of the Supreme Court of Canada have frequently relied in developing their approaches to s 2(b) (see e.g. Justice McLachlin's reasons for judgment in Keegstra, supra note 9 at 805), holds strongly to the view that only those expressive activities that can be said to further one or more of the rationales for protecting free speech should receive the protection of the Free Speech clause of the First Amendment (see his book, Free Speech: A Philosophical Enquiry, supra note 10, at 101-06). He makes the same argument in relation to s 2(b) of the Charter in his recent article, "Expression and Its Consequences" (2007) 57 UTLJ 705.
(68) Keegstra, supra note 9.
(69) R v Zundel, supra note 20. Richard Moon has used the word "shocked" to describe his reaction to the Court's decision to afford constitutional protection to such expressive activity; see supra note 3 at 434-35.
(70) David Lepofsky, supra note 3 at 49 et seq also expresses concern about the range of expressive activities protected under the Irwin Toy approach in his critique of the decision. However, he offers no alternative approach to defining the reach of freedom of expression to the one taken by the Court. Several other scholars have argued for a narrowing of the scope of s 2(b); see e.g. K Mahoney, "R. v Keegstra: A Rationale for Regulating Pornography?" (1992) 37 McGill LJ 242; and I Lee, "Can Economics Justify the Constitutional Guarantee of Freedom of Expression?" (2008) 21 Can JL & Juris 355.
(71) Irwin Toy, supra note 1 at 968.
(72) Handyside, supra note 29, cited ibid at 969.
(73) Big M Drug Mart Ltd, supra note 43 at 344.
(74) Peter Hogg makes the latter argument in Chapter 36.8(c) his textbook, Constitutional Law of Canada, loose-leaf (consulted on 28 March 2011), (Toronto: Thomson Carswell)
(75) J Bakan et al, Canadian Constitutional Law, 4th ed (Toronto: Emond Montgomery, 2010) at 3.
(76) See supra note 45 and accompanying text.
(77) Supra note 9 at 733-34.
(78) Sensitivity to context has been an important feature of the Supreme Court's approach to s 1 since Edmonton Journal, supra note 15.
(79) See e.g. Smith v Canada (AG), 2001 SCC 88,  3 SCR 902.
(80) See e.g. R v Delaronde,  1 SCR 213, 115 CCC (3d) 355.
(81) See e.g. R v Smith,  2 SCR 1120, 52 CCC (3d) 97.
(82) See e.g. R v Pearson,  3 SCR 665, 777 CCC (3d) 124.
(83) See e.g. R v Smith,  1 SCR 1045, 40 DLR (4th) 435.
(84) See e.g. R v Heywood,  3 SCR 761, 120 DLR (4th) 438.
(85) See e.g. Reference Re Prov Electoral Boundaries (Sask),  2 SCR 158, 81 DLR (4th) 16.
(86) (Minister of Employment and Immigration),  1 SCR 497, 170 DLR (4th) 1.
(87) I am not averse in principle to making further refinements to our understandings of these rationales; I'm simply suggesting that none is required in order to deal with these two content-based kinds of expression.
(88) Keegstra, supra note 9 at 830.
(89) 458 US 747, 102 S Ct 3348 (1982).
(90) Ibid at 763 -64 [emphasis added].
(91) Gertz v Robert Welch, Inc, 418 US 323, 94 S Ct 2997 at 3 40 (1974). It should be noted that Justice Powell was careful to distinguish between false statements of fact and false opinions; while the former were not entitled to constitutional protection, the latter were.
(92) Chaplinsky v New Hampshire, 315 US 568, 62 S Ct 766 (1942).
(93) Miller v California, 413 US 15, 93 S Ct 2607 (1973).
(94) Central Hudson Gas & Electric Corp v Public Service Commission of New York, 447 US 557, 112 S Ct 2343 at 563-64 (1980) [Central Hudson].
(95) RAV v City of St Paul, 505 US 377, 112 S Ct 2538 at 388 (1992).
(96) Central Hudson, supra note 94, at 566.
(97) It is true, as one of the external reviewers pointed out, that the practice of definitional balancing under the free speech clause has been criticized by some scholars in the United States. It is also true that that criticism has resulted in some formerly unprotected categories of speech-notably commercial expression and defamatory speech directed at public officials and public figures-being moved by the United States Supreme Court in recent decades into the realm of protected categories. However, some content-defined categories of speech remain unprotected as a result of definitional balancing, and it is still the case that freedom of expression is afforded a higher degree of protection in the United States than in any other country.
(98) Supra note 43 at 331-36.
(99) Ibid at 331 [emphasis added].
(100) RSC 1970, c L-13, s 4.
(101) Supra note 43 at 331.
(102) R v Edwards Books and Art Ltd,  2 SCR 713, 35 DLR (4th) 1 [Edwards Books cited to SCR].
(103) RSO 1980, c 453.
(104) Supra note 102 at 752.
(105) Big M Drug Mart, supra note 43 at 334 [emphasis added].
(106) Ibid at 351 [emphasis added].
(107) Ibid at 331 [emphasis added].
(108) Ibid at 362 [emphasis added].
(109) It should be noted that the Court has made pronouncements to the same or similar effect at the s 1 stage in two of its s 2(b) cases, R v Butler, supra note 19 and R v Zundel, supra note 20. For example, in Butler, supra note 19, Justice Sopinka, writing for six members of the Court, held that, "to impose a certain standard of public and sexual morality ... is inimical to the exercise and enjoyment of individual freedoms, which form the basis of our social contract ..., [and therefore] is not a legitimate objective [for the purposes of s 1]."
(110) Irwin Toy, supra note 1 at 978.
(111) These two goals were said to have been "to prevent the pain suffered by target group members and to reduce racial, ethnic and religious tension in Canada ...": Keegstra, supra note 9 at 758.
(112) Ibid at 730. It should perhaps be noted that the "objective" to which he refers in this passage is the objective of the purveyor of hate propaganda, not the government.
(113) Ibid at 973 [emphasis added].
(114) Ibid at 973-74 [emphasis added].
(115) There is a helpful discussion of the delicate nature of judicial inquiries into allegations of bad faith on the part of governments in the context of the First Amendment free speech cases in Tribe, supra note 49 at 591-94.
(116) Supra note 43.
(117) These connections are evident in both of the following well-known texts, John Milton, Complete English Poems, of Education, Areopagitica, 4th ed by Gordon Campbell (London: JM Dent & Sons, 1990) and John Locke, A Letter Concerning Toleration (Buffalo, NY: Prometheus, 1990) (1st ed 1689).
(118) Richard Moon is critical of this distinction for many of the same reasons I provide here; see Moon, supra note 3 at 436 et seq. For other critiques of the distinction, see Macklem, supra note 3; K Kowal, "The First Amendment and Section 2(b) of the Charter: Predicting the Outcome of the Tobacco Products Advertising Ban in Canada" (1991) 1 Media & Comm L Rev 237; and J Cameron, "A Bumpy Landing: The Supreme Court of Canada and Access to Public Airports Under s 2(b) of the Charter' (1991) 2 Media & Comm L Rev 91.
(119) Jamie Cameron makes this point in her critique of Irwin Toy. She is also critical of the distinction, but for different reasons than mine; in her view, claimants should never be required to connect their expressive activities to the rationales for protecting freedom of expression. She also argues that courts should limit their inquiry at the infringement stage to the effect of the impugned governmental action (supra note 3 at 271-75).
(120) Supra note 11.
(121) My colleague, Margot Young, has suggested that the distinction might reflect a sense on the Court's part that, because governments will sometimes have difficulty predicting the consequences on expressive activities of the actions they take, they are entitled to a somewhat more sympathetic response when those actions are challenged on the basis of their effects alone. According to this theory, the additional burden placed on challengers in such cases is designed to give expression to that enhanced sympathy. Whatever one thinks about the merits of such a theory, it is not one that the Court has ever seen fit to articulate.
(122) See e.g. Lepofsky, supra note 3 at 66.
(123) The Court cites the following publications of these two scholars: A Cox, Freedom of Expression (Cambridge, Mass: Harvard University Press, 1981) and T Scanlon, "A Theory of Freedom of Expression," in RM Dworkin, ed, The Philosophy of Law (London: Oxford University Press, 1977).
(124) For a good summary of this feature of American law in this area, see Tribe, supra note 49, ch 12-2.
(125) Justice McLachlin acknowledged in her concurring reasons for judgment in Commonwealth of Canada, supra note 120 at 248, that there is support in some quarters for the proposition that content-neutral infringements should be less difficult for governments to justify than content-based infringements; however, she expressed some misgivings about the merits of that proposition, and it played no meaningful role in her application of the Oakes framework in that case. Here is what she said about its use at the s 1 stage:
[C]are must be taken to avoid the trap of acceding to limits on expression on the basis that they relate to content-neutral consequences rather than content. Denial of a particular time, place or manner of expression regardless of content may effectively mean denial of the right to communicate. Conversely, as abhorrent as arbitrary or unfair content-related limitations may be, it must be conceded that when carefully tailored they may be integrally tied to important government purposes outweighing any interest a speaker may have in communicating a conflicting message. Restrictions on content may be capable of being justified on the basis that certain messages are incompatible with the purpose or function of a particular government institution. Pornography or literature promoting drug use, for example, might be legitimately banned at a school or a children's festival in a government park. The point is that generalizations are of little assistance. What is essential is that the court in each case undertake the process of balancing and weighing the true interests at stake with a view to determining whether the limit on free expression in question is "reasonable" and "demonstrably justified in a free and democratic society."
There is very little if anything in the Court's s 2(b) jurisprudence generally to suggest that the distinction is seen by the Court to have any significance at the s 1 stage. It is noteworthy in this regard that, in the same time period in which it upheld content-based restrictions on advertising to children (Irwin Toy), hate propaganda (Keegstra), and obscenity (Butler), the Court struck down two content-neutral restrictions on access to public property for expressive purposes (see Commonwealth of Canada and Ramsden v Peterborough, supra note 11).
(126) (Cambridge, Mass: Harvard University Press, 1990).
(127) Ibid at 15.
(128) See e.g. "The Meaning of Symbolic Speech under the First Amendment" (1973) 21 UCLA L Rev 29.
(129) Supra note 49, ch 12-2.
(130) Supra note 123.
(132) See e.g. "Flag Desecration: A Case Study in the Roles of Categorization and Balancing in First Amendment Analysis" (1975) 88 Harv L Rev 1482.
(133) Supra note 126 at 17.
(134) Ibid at 18.
(135) Ibid at 21.
(136) Ibid at 22-23.
(137) Ibid at 23-24.
(138) Ibid at 24-26.
(139) American free speech jurisprudence provides numerous examples of cases in which the appropriate categorization was far from obvious-see e.g. City of Renton v Playtime Theatres, Inc, 475 US 41, 106 S Ct 925 (1986); Boos v Barry, 485 US 312, 108 S Ct 1157 (1988); and Simon & Schuster, Inc v Members of the New York State Crime Victims Board, 502 US 105, 112 S Ct 501 (1991).
(140) Of the people who have already commented on the Irwin Toy framework, only David Lepofsky has made reference to this feature of it. He, too, sees it as a significant flaw: supra note 3 at 78-80.
(141) See e.g. United States v National Treasury Employees Union, 513 US 454, 115 S Ct 1003 (1995).
(142) See e.g. Speiser v Randall, 357 US 513, 78 S Ct 1332 (1958) and Rust v Sullivan, 500 US 173, 111 S Ct 1759 (1991).
(143) See e.g. Bantam Books, Inc v Sullivan, 372 US 58, 83 S Ct 631 (1963) and Meese v Keene, 481 US 465, 107 S Ct 1862 (1987).
(144) Supra note 43 at 336.
(145) Ibid at 336-37.
(146) Supra note 102 at 759.
(149) Supra note 5 at para 32.
(150) 2006 SCC 6,  1 SCR 256 at para 148.
(151) The reason the Court was able to use the Irwin Toy framework in this case in spite of this fact was that the Court viewed the case as being analogous to the access to public property for expressive purposes kind of case, and, as indicated supra note 11, the Court has attempted to merge its approach to that kind of case into the Irwin Toy framework.
(152) Supra note 5 at para 31.
(153) Ibid at para 5.
(154) Ibid at para 36. Ironically, the Court cites Irwin Toy in support of this requirement. It should be noted that the understanding of those rationales that the Court employs here appears to be a relatively narrow one. The cited passage refers only to "free expression on matters of public or political interest," suggesting that the Court is limiting itself to the self-government and search for truth rationales, and ignoring the self-realization rationale. A proper application of the purposive approach would have included reference to all three rationales.
(155) Greater Vancouver Transportation Authority v Canadian Federation of Students, 2009 SCC 31  2 SCR 295 at para 28 [emphasis in original].
Robin Elliot, Professor of Law, University of British Columbia. The author is grateful to Bill Black, Susan Chapman, Peter Hogg and Margot Young for their helpful comments on earlier drafts of this paper. He is also grateful to his research assistants, Keith Evans, Colin Trehearne, and Brandon Langhjelm, and to Andrew Irvine for their valuable input. Any errors the paper is found to contain are, of course, the sole responsibility of the author.