Twitter Blocking, Freedom of Expression, and Public Forums

Canadian legal twitter and podcasting celebrity Emilie Taman, along with a few other plaintiffs, have started a constitutional challenge in which they allege that Ottawa Mayor Jim Watson has violated their freedom of expression by ‘blocking’ them on his twitter account. As described by the Ottawa Citizen’s David Reevely:

When Mayor Jim Watson blocks people on Twitter he’s violating their constitutional rights, a trio of Ottawa activists says, and they’re going to court to try to make him stop.

The case is the first of its kind in Canada, says human-rights lawyer Paul Champ, who’s representing them…

[The Plaintiffs] all say Watson has cut them off from his Twitter feed after they’ve annoyed him. Which is not OK, they argue, because the mayor is a public official who uses his Twitter account for public purposes, to communicate public information and explain things he’s doing as Ottawa’s top civic politician.

The claim has raised some eyebrows. In particular, political scientist and constitutional expert Professor Emmett Macfarlane commented on twitter that “(i)f this [challenge] succeeds then rights don’t have any meaning anymore”.  Those are strong words, against which I want to push back a bit in this post.

I have not yet read the Plaintiffs’ statement of claim. But as I see it, the key to the argument is to view a public official’s twitter account, to the extent that it is regularly used as the account of a public official for public purposes, as a sort of “public forum”, from which individuals cannot be unreasonably or arbitrarily excluded.  It is something like this argument that was successful in the similar suit against Donald Trump, and it is the one I want to address here.

The idea of a public forum is a staple of US First Amendment jurisprudence. It recognizes that there are certain forums for expressive activity where it would be fundamentally contrary to a free society to permit the exclusion of individuals on the basis of the viewpoints they express. The basic idea was famously expressed by Professor Kalven in these terms:

[I]n an open democratic society the streets, the parks, and other public places are an important facility for public discussion and political process. They are in brief a public forum that the citizen can commandeer; the generosity and empathy with which such facilities are made available is an index of freedom.

Cited in Committee for the Commonwealth of Canada v. Canada, [1991] 1 SCR 139.

Different categories have since been created – traditional public forums (e.g. public parks, sidewalks, etc.), designated or limited public forums (e.g. university spaces or auditoriums used for public purposes), and non-public forums. In the US constitutional law context, these designations result in different degrees of scrutiny in terms of justifying a restriction on access, with access to traditional public forums attracting the highest level of scrutiny, and non-public forums attracting the least.

While the “public forum” analysis as such is a unique feature of US constitutional law, Canadian courts have grappled with similar issues, and reached similar conclusions. A few cases come to mind. In Committee for the Commonwealth of Canada v. Canada, certain individuals were prohibited from soliciting and leafleting in an airport that was government owned and controlled. There were six (!) sets of reasons, so I will not dare hazard a summary of the ratio of the decision. But suffice it to say that the Court held in favour of the leafletters, on the basis that the location was and should be available for expressive activities.

Perhaps more on point, in Greater Vancouver, the Supreme Court held that a public bus service that provided advertising space could not prevent individuals from advertising on the bus without a compelling section 1 justification. According to the Court:

The very fact that the general public has access to the advertising space on buses is an indication that members of the public would expect constitutional protection of their expression in that government‑owned space. Moreover, an important aspect of a bus is that it is by nature a public, not a private, space. Unlike the activities which occur in certain government buildings or offices, those which occur on a public bus do not require privacy and limited access… Like a city street, a city bus is a public place where individuals can openly interact with each other and their surroundings. Thus, rather than undermining the purposes of s. 2(b), expression on the sides of buses could enhance them by furthering democratic discourse, and perhaps even truth finding and self‑fulfillment.

Greater Vancouver Transportation Authority v. Canadian Federation of Students — British Columbia Component, 2009 SCC 31 at para 43.

Canadian cases have extended this type of conclusion to other public forums, such as sidewalks and parks: see generally the cases cited in Bracken v. Niagara Parks Police, 2018 ONCA 261 at para 39-44. As Mr. Justice Miller put the point in Bracken, at para 49, in relation to a protestor in a public marketplace:

… Grandview Plaza is a place where people congregate and must expect to interact with others. That is precisely what made it an attractive destination for Mr. Bracken. Nothing that happens there requires quiet or an absence of distraction. Indeed, neither quiet nor the absence of distraction is even possible there. As in Greater Vancouver,

[u]nlike the activities which occur in certain government buildings or offices, those which occur [in the Parks] do not require privacy and limited access … Like a city street, [the Parks are] a public place where individuals can openly interact with each other and their surroundings (Greater Vancouver, at para. 43, emphasis added).

The point of these cases is not that one has a “right” to advertise on buses or to access any other particular “platform” for expression, as standalone proposition. The point is rather that if the Government creates expressive opportunities ostensibly open to all, it cannot unreasonably or arbitrarily prevent individuals from using those opportunities.  In my view, we should be particularly concerned when they do so to suppress certain messages they find distasteful generally, or critical of the public official or government entity specifically.

Is a public official’s twitter account properly considered a public forum of some sort, such that some degree of constitutional scrutiny should apply where individuals are excluded from participating in that forum? I think there are good reasons to say it should be.

First, blocking an individual from seeing and responding to tweets in the twitter thread created by the public official deprives that individual of their freedom to express themselves in a particularly important way, in a particularly effective forum, and moreover, in a forum explicitly designed for those types of communications. The critical importance of unfettered speech on issues of public interest, and speech critical of public officials and government in particular, is too widely accepted to warrant a citation, but here’s one anyway:

“The freedom of individuals to discuss information about the institutions of government, their policies and practices, is crucial to any notion of democratic rule.  The liberty to criticize and express dissentient views has long been thought to be a safeguard against state tyranny and corruption.

Canadian Broadcasting Corp. v. New Brunswick (Attorney General), [1996] 3 SCR 480 at para 18.

Second, and relatedly, blocking dissenters prevents many other twitter users from seeing the critical or other commentary of the speaker in this uniquely effective forum, and engaging with that commentary in turn. As the Court has recognized, freedom of expression protects “listeners as well as speakers”, and we should be wary of state action that interferes with either (see e.g. Ford v. Quebec (Attorney General), [1988] 2 S.C.R. 712, at pp. 766-67; Edmonton Journal v. Alberta (Attorney General), [1989] 2 S.C.R. 1326 at 1339-40).

In my view, curating a twitter feed to effectively block dissenting or critical voices, in an open and public venue perfectly compatible with such reactions and commentary, undermines both of these important constitutional values.

It is in my view not an adequate response to say that the blockee could create a new twitter account, and therefore could still read the tweets of the blocker. While this is true, it is not responsive to the unique constitutional harm engaged. To me, the harm is not that the individual cannot read the statements or commentary of the public figure, but rather that they cannot express themselves and participate in an especially critical forum created by that public official, i.e., the very twitter feed in which the public announcement or statement is made. The inconvenient workaround of creating a separate twitter account to view the statements does not solve the problem of the arbitrary exclusion from the public forum designed for expression and debate, and the uniquely important expressive opportunities that the public official has created.

All of which is to say that I think there is a strong argument, grounded in existing constitutional law and in the fundamental purposes of freedom of expression, for considering a publicly available twitter account used for public purposes by a public official to be the type of forum in which restrictions on access may deserve constitutional scrutiny. The more twitter and other social media accounts become the predominant method for communications with and interactions between elected officials and their constituents, increasingly in preference to stump speeches and news conferences of previous generations, the stronger this argument becomes.

Although other analogies to a public official’s twitter feed have been floated, the most persuasive from my perspective would be the making of a government announcement on a public street or a public park, or in a government owned or rented building or auditorium made open to members of the public generally.

In those circumstances, I think we would consider it a rather intolerable intrusion on freedom of expression and assembly if the public official, at a function ostensibly available to all members of the public, sent around police officers or security guards to evict those do not seem to support the public official or their message. This would not only trammel upon the freedom of expression of the excluded dissenters, but would deprive listeners and viewers of being exposed to contrary viewpoints. Perhaps as concerning, from my perspective, the process of regularly blocking dissenters while permitting supporters to access the forum creates the false impression of unanimous support for the public official’s statement in the forum in which that statement is made. This is of course a common tactic in certain regimes, but not one normally seen in free and democratic societies.

Yes, in such situations – as in the context of a twitter blocking – the individuals excluded from the public venue can express themselves elsewhere. You can go down to another public area or find a private one, where no one is actually located to hear your message (a free speech zone, perhaps?). But in my view, that does not eliminate the constitutional issue. The key feature and logic of public forums is that they create a particularly effective venue for expressive activities, both in terms of a larger audience generally, and in terms of reaching an audience who have an interest in the public official or the content of their speech. Saying that an individual is free to express themselves in this uniquely effective forum if they support the public official, and in other far less effective forums if they do not, does not seem to address that concern.

And contrary to Professor Macfarlane’s view, I think there is a principled basis to say that ‘muting’ does not raise the same constitutional issues. That is because muting does not prevent individuals from accessing and participating in this modern public forum, nor does it prevent anyone else from benefiting from their commentary in that same forum. The point is not that you have a right to be listened to by public officials or anyone else, but rather that you cannot be unreasonably excluded from public forums made available for that purpose. Thus, muting strike me as the rough equivalent of a public official who decides to give a speech or hold a public event that permits dissenting voices, but then averting their eyes from critical signs or ignoring contrary speech within earshot. That is ok.

Public officials can of course avoid this issue entirely, by not creating the public or quasi-public forum in the first place. They can create a ‘private’ or ‘protected’ twitter account, for instance, in which they screen and only accept followers and commenters who support their viewpoint, and curate that list as aggressively as they like. This would be like holding an exclusive private event or conference limited to party members, in which property rights are used to exclude those from the conversation that the public official does not want interact with or hear from. In other words, it would not be a “public” and open forum at all, but a “private” and exclusive one.

However, having chosen to create or utilize a public forum to reach and engage with the widest possible audience, and at least ostensibly making it open to all to hear the views of the politician and express views in return, it strikes me as constitutionally problematic for a public official to require that forum, which they exclusively control, to be used only to praise and cheer the public official, and not to criticize or question her or him.

Finally, I would add that it does not strike me as problematic to block twitter users whose commentary can be properly characterized as threatening, abusive, harassing, or unduly disruptive, just as such individuals could be excluded from a more traditional or designated public forum for those reasons. But that would be a reasonable justification for a restriction on speech, not a justification for taking it outside of the constitutional arena altogether.

There are of course many potential wrinkles here. It may be difficult to determine whether a public official has created and used a twitter account for public or private purposes, and what to do if it is a little of both. There are also practical obstacles that the claimants may need to traverse, such as a Court’s fear that giving effect to the constitutional arguments in this context will create a flood of twitter litigation. A court might even reject the premise altogether, and treat a public official’s twitter feed as less the “online equivalent of government property”, as Andrea Gonsalves and Justin Safayeni have described it, and more the online equivalent of an personal diary, over which the official has and should have absolute discretion.  As with any novel case, I’m sure there are plenty of other issues that will need to be hashed out, and at the end of the day, Professor Macfarlane’s doubts may be vindicated. But the claim strikes me as being perfectly viable from a constitutional law perspective, even if it is not ultimately successful.

I will conclude with this.  For all its incurable faults, the great promise of social media is that it can open up new and uniquely democratic forums for public engagement, including public dissent and criticism. Unlike public speeches that are only available to those living nearby, or news releases and no-questions press conferences that are decidedly unidirectional, twitter creates a forum for constant engagement, debate, interaction and feedback. It provides citizens – particularly those with relatively little political, economic or social power – a meaningful opportunity to have their views broadcast and heard, in nearly equal measure to the public figures they support or denounce. In my view, excluding people from that conversation on the basis of their political opinions or substantive viewpoints, in a modern forum designed deliberately for the purpose of fostering that type of dialogue, is worth scrutinizing through a constitutional lens.

Charitable Status and Freedom of Expression: Testing Labour Union Exceptionalism in the context of the Charter’s Fundamental Freedoms

The charitable organization Canada Without Poverty (“CWP”) has created some buzz lately with its constitutional challenge to a provision in the Income Tax Act that makes charitable tax status contingent on refraining from engaging in certain “political activities”.

As a preliminary matter, there is always a risk in assessing laws impacting expression that our thinking will become clouded by sympathies for the expression at issue. In order to avoid this at the outset of this egregiously long post, I would like to invite readers to not think of this as a challenge brought by an anti-poverty group. No one likes poverty. Think of it as a challenge brought by an organization whose otherwise lawful political expression you find obnoxious or distasteful. I, for instance, will imagine that the challenge was brought by a not-for-profit organization dedicated to promoting the interests of self-described “foodies”.

So the question is: does this lawful organization (whose ideas or objectives you dislike) have a constitutional entitlement to favourable tax treatment not available to other organizations, and to use the additional funds for their political purposes, as an incident of their fundamental freedoms?

Positive and Negative Rights

This challenge raises a number of unresolved issues that go to the very nature of the concept of “freedom” used in section 2 of the Charter, and in particular everyone’s favourite but murky (if not analytically unstable) distinction of “positive” vs. “negative” rights.

These difficult issues arise because the Government does not prohibit, restrict, or otherwise impose sanctions on organizations for engaging in political activities or expression. No one is stopping any organization from saying or doing anything, as such.

Rather, what the law does is make beneficial tax status contingent on refraining from engaging in political activities, including political expression. As I understand it, charitable organizations can engage in political activities and expression, or obtain tax breaks, but not both.

The difficulty with the CWP’s position is that we normally think of “freedom” as requiring the government to not interfere with the fundamental freedoms (religion, expression, assembly and association), but not as requiring positive state assistance for those activities. Presumably, we would all have greater opportunity to expend funds on religious, expressive or associational activities if we were afforded state assistance for them, be it through beneficial tax status, government grants, or positive legislative protections designed to facilitate these activities. But if you walk into court and say that the government has violated your fundamental freedoms because they have not left you with enough post-tax income to build a church or to run prime time election ads or to rent lane-time so your bowling association can practice, you will probably not get very far.

This point is not lost on the lawyers for the CWP. In their Notice of Application, they assert that they are seeking an entitlement to charitable status as such, but rather take the existence of charitable tax status as a given:

  1. CWP is not arguing that Parliament is constitutionally obliged to confer the benefit of charitable status for the promotion of any particular purpose or view. Though an argument could be advanced that governments have an obligation to provide statutory or financial support for organizations such as CWP to promote the relief of poverty, that is not the issue in this case. Parliament has accepted that relief of poverty is a charitable purpose and CWP has been granted charitable status to pursue this purpose. CWP relies on the fact that even if there is no constitutional obligation to provide charitable status for the relief of poverty, Parliament must ensure that where it chooses to provide the benefit, it does so in a manner that complies with the Charter. Restrictions imposed on CWP’s political expression must therefore be in compliance with the Charter. Section 149.1 (6.2) has as its clear purpose the restriction of political activities or expression. All of CWP’s activities that are subject to this restriction have expressive content, thereby bringing them prima facie within the scope of s. 2(b) protection.

While there are a number of ways this challenge could go,* I will focus on the constitutional principle underlined above – that while the government may not be constitutional required to confer a certain benefit, once it chooses to do so, it must do so in compliance with the Charter.

This principle is rather obviously true in general, but is more readily applicable in certain respects than in others. In particular, in the context of section 15 equality challenges, the government cannot extend a benefit to some and deny it to others on discriminatory grounds, and then claim it has not breached the Charter because the persons deprived of the benefit would have no entitlement to it in addition to their right to not be unlawfully discriminated against. The whole point of equality rights – and in particular the rights to “equal benefit” and “equal protection” of the law – is to forbid the discriminatory extension of benefits, burdens and protections.

The question the CWP application raises is different.  It is not raising a section 15, relative-entitlement equality claim – I am entitled to this state benefit/protection/support because others get it.  Rather, by relying on section 2, the CWP claim (or at least the angle I am focussing on) enters into the field of absolute entitlements – I am entitled to this additional state benefit/protection/support regardless of what other people get, because it is necessary to permit me to meaningfully exercise my fundamental freedoms.

Positive and Negative Rights and the Fundamental Freedoms

Generally, the courts have been resistant to extend state benefits or protections in that way under the fundamental freedoms. For instance, in Haig v. Canada, a case involving the government excluding persons from voting in a referendum due to their fluid residency status, L’Heureux-Dube J. made the following remarks:

As a starting point, I would note that case law and doctrinal writings have generally conceptualized freedom of expression in terms of negative rather than positive entitlements. (…)

It has not yet been decided that, in circumstances such as the present ones, a government has a constitutional obligation under s. 2(b) of the Charter to provide a particular platform to facilitate the exercise of freedom of expression.  The traditional view, in colloquial terms, is that the freedom of expression contained in s. 2(b) prohibits gags, but does not compel the distribution of megaphones… [at 1035; emphasis added]

In applying this principle to the expressive ‘benefit’ at issue (the ability to express oneself through voting in a referendum), the Court in Haig found that there was no such entitlement:

A referendum is a creation of legislation.  Independent of the legislation giving genesis to a referendum, there is no right of participation.  The right to vote in a referendum is a right accorded by statute, and the statute governs the terms and conditions of participation.  The Court is being asked to find that this statutorily created platform for expression has taken on constitutional status.  In my view, though a referendum is undoubtedly a platform for expression, s. 2(b) of the Charter does not impose upon a government, whether provincial or federal, any positive obligation to consult its citizens through the particular mechanism of a referendum.  Nor does it confer upon all citizens the right to express their opinions in a referendum.  A government is under no constitutional obligation to extend this platform of expression to anyone, let alone to everyone.  A referendum as a platform of expression is, in my view, a matter of legislative policy and not of constitutional law.

The following caveat is, however, in order here.  While s. 2(b) of the Charter does not include the right to any particular means of expression, where a government chooses to provide one, it must do so in a fashion that is consistent with the Constitution.  The traditional rules of Charter scrutiny continue to apply. Thus, while the government may extend such a benefit to a limited number of persons, it may not do so in a discriminatory fashion, and particularly not on ground prohibited under s. 15 of the Charter. [at 1041; emphasis added]

On this logic, the Court has rejected various constitutional challenges where the claimant sought state assistance or a certain ‘platform’ to facilitate their expression (i.e. a positive right). For instance, in NWAC, the Court rejected the claim that the government was constitutionally required under s. 2(b) to provide an aboriginal women’s group with funding and access to facilitate their position in constitutional negotiations:

It cannot be claimed that NWAC has a constitutional right to receive government funding aimed at promoting participation in the constitutional conferences (…) The respondents conceded as much in paragraph 91 of their factum as well as in oral argument. Furthermore, the provision of funding and the invitation to participate in constitutional discussions facilitated and enhanced the expression of Aboriginal groups. It did not stifle expression. (…)

At this point, I should add that it cannot be said that every time the Government of Canada chooses to fund or consult a certain group, thereby providing a platform upon which to convey certain views, that the Government is also required to fund a group purporting to represent the opposite point of view.  Otherwise, the implications of this proposition would be untenable.  For example, if the Government chooses to fund a women’s organization to study the issue of abortion to assist in drafting proposed legislation, can it be argued that the Government is bound by the Constitution to provide equal funding to a group purporting to represent the rights of fathers?  If this was the intended scope of s. 2(b) of the Charter, the ramifications on government spending would be far reaching indeed. [at 654-656; emphasis added]

At one time, this was the default assumption that applied to all of the fundamental freedoms. As Robert Charney has recently explained, it has been applied in the context of freedom of religion, in cases like Adler:

Persons seeking funding for private religious schools have argued that without government funding they are unable to establish a religious school, or, if established, students who might want to attend would be unable to do so because they could not afford the tuition. In other words, they argued, that for at least some individuals, the right to attend a private religious school was meaningless in the absence of government funding to build and support such schools. This argument was rejected by the Supreme Court, which held that freedom of religion does not entitle one to state support for one’s religion. As Chief Justice Dubin stated in the Ontario Court of Appeal:

The right [to freedom of religion under Charter s.2(a)] involves the freedom to pursue one’s religion or beliefs without government interference, and the entitlement to live one’s life free from state-imposed religious beliefs. It does not provide . . . any entitlement to state support for the exercise of one’s religion.

Robert E. Charney, “Should the Law society of Upper Canada Give Its Blessing to Trinity Western University Law School” (2015) 34 NJCL 173 at 182; see also Adler, at para 199-200, per McLachlin J, at paras 171, 175, per Sopinka J., and at para 58, per L’Heureux‑Dubé J..

Similarly, in the context of freedom of association, the Court (at one time) rejected the proposition that the Government was required to extend certain ‘positive’ entitlements under the rubric of the fundamental freedoms, such as affirmative labour rights designed to facilitate or promote associational activities.

In Delisle v. Canada, for instance, the Court rejected the submission that excluding certain employees (there, RCMP members) from the protections found in the general labour relations statute violated their freedom of association. The RCMP members were left free to associate and to make representations to their employer about working conditions; they simply were not provided with the affirmative statutory protections necessary to enhance the power of that association in the collective bargaining context. In essence, the employer could ignore them. The majority explained the distinction between the principle’s application in the context of section 15 and in the context of the fundamental freedoms:

The structure of s. 2 of the Charter is very different from that of s. 15 and it is important not to confuse them.  While s. 2 defines the specific fundamental freedoms Canadians enjoy, s. 15 provides they are equal before and under the law and have the right to equal protection and equal benefit of the law.  The only reason why s. 15 may from time to time be invoked when a statute is underinclusive, that is, when it does not offer the same protection or the same benefits to a person on the basis of an enumerated or analogous ground (on this issue, see Schachter v. Canada, [1992] 2 S.C.R. 679), is because this is contemplated in the wording itself of s. 15.  The distinguishing feature of s. 15 is that the Charter may require the government to extend the special status, benefit or protection it afforded to the members of one group to another group if the exclusion is discriminatory and is based on an enumerated or analogous ground of discrimination. (…)

It is because of the very nature of freedom that s. 2 generally imposes a negative obligation on the government and not a positive obligation of protection or assistance. (…)

On the whole, the fundamental freedoms protected by s. 2 of the Charter do not impose a positive obligation of protection or inclusion on Parliament or the government, except perhaps in exceptional circumstances which are not at issue in the instant case.  In accordance with the decision of the majority of this Court in Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), supra, there is no violation of s. 2(d) of the Charter when certain groups of workers are excluded from a specific trade union regime.  The ability to form an independent association and to carry on the protected activities described below, the only items protected by the Charter, exists independently of any statutory regime. (…) [at paras 25, 26, 33]

As can be seen, the Court was characteristically cautious to never say never. In Haig, for instance, the majority mused that “a situation might arise in which, in order to make a fundamental freedom meaningful, a posture of restraint would not be enough, and positive governmental action might be required”. It gave the example of “legislative intervention aimed at preventing certain conditions which muzzle expression, or ensuring public access to certain kinds of information”, which it implied may require constitutional protection, despite the fact that it could be characterized as an assertion of a ‘right’ to positive protection or assistance (Haig at 1039). And as discussed below, the Court has taken this ball and thrown deep, especially (or perhaps exclusively) in the context of freedom of association and labour rights.

However, even after that ball started rolling, the Court has generally hewed closely to the orthodox distinction in most other contexts. In Siemens v. Manitoba, the Court again confirmed that freedom of expression did not entitle an individual to vote in a referendum, relying on Haig. And in Baier v. Alberta, the Court developed a somewhat obscure framework for the extension of positive rights in the context of section 2, in the course of rejecting the claim that freedom of expression was violated by excluding certain persons from running in a school board election. Notwithstanding the unique expressive advantage conferred by being a school trustee – i.e. it enhances the meaningfulness of expression – the court found that “claiming a unique role is not the same as claiming a fundamental freedom” (Baier at para 44).

Thus, I take it to be the general rule under section 2 that state may not act to impede religious belief or practice, expression, or associational activities, but it need not actively facilitate, promote, enhance or assist those activities.

Departing from the Rights vs. Freedom Distinction

The Court has since departed from this general s. 2 rule in a rather big way, albeit almost exclusively in one particular context: labour rights designed to facilitate meaningful association in the workplace. The first departure came in Dunmore v. Ontario, where the Court found that section 2(d) of the Charter required the extension of statutory rights specifically designed to facilitate the act of association, namely, protections against unfair labour practices of employers discriminating against employees who choose to associate (i.e. firing someone who joined a union), which employers are free to do at common law.

Since Dunmore, proceeding to BC Health and Fraser, and on to MPAO and Saskatchewan Federation of Labour, the Court has found that a broader range of affirmative statutory rights are necessary to permit workers to “meaningfully” exercise their fundamental freedom of association in the labour relations context. In these cases, no law or state actor stopped anyone from freely associating, generally speaking. The problem was that the association so created was not able to achieve what it was designed to achieve in the absence of unique statutory protections: i.e. engaging in meaningful collective bargaining, including by placing an obligation on the employer to bargain in good faith, protecting employees from termination lawful at common law, having access to a right to strike, and so on.

I do not think the Court has yet extended this principle outside the labour relations context, with one possible exception. In CLA, the Court found that section 2(b) of the Charter could require the government to disclose information in its possession where doing so was necessary to facilitate expression on the subject matter of the disclosure.  I happen to think is not so much an exception to the general rule, but that discussion is beyond the scope of this post. In any event, and beyond the CLA case, the extension of fundamental freedoms to require positive state support appears to only really apply in the context of labour relations. As the Court stated in Dunmore, with perhaps some degree of understatement, “it may be asked whether the distinction between positive and negative state obligations ought to be nuanced in the context of labour relations” (at para 20).

Notably, in defending the position that section 2 imposes positive obligations on the government in some circumstances, the Court has arguably gone beyond the wholly defensible position that the distinction between (positive) rights and (negative) freedoms can be murky and will at least admit of borderline cases, to the more radical proposition that there is really no distinction at all, stating for instance that “(t)he freedom to do a thing, when guaranteed by the Constitution interpreted purposively, implies a right to do it.” (Fraser, at para 67).

I think that taken too far, this view is problematic, if for no other reason that it would tend to put the courts in the position of doling out governing funding and statutory rights based on some arbitrary baseline entitlement to ‘meaningful freedom’. I doubt that is something that would be contemplated outside of the unique labour relations context. As I have put it elsewhere:

For example, consider freedom of religion. Section 2(a) does not impose any positive duty on the government, even if my lack of resources makes an important incident of this freedom (e.g., going on pilgrimages to Mecca or building a church) all but illusory, vapid or ‘impossible to exercise’. Likewise, if you are without the means or opportunity to effectively distribute your message to an audience, or simply no one cares to listen to you, your freedom of expression may be effectively rendered ‘pointless’. The absence of state action in this case may have the effect of ‘precluding’ meaningful expression, and the futility of the enterprise may indirectly ‘discourage’ it, but this does not entitle you to a constitutional remedy on the basis of government inaction. (…)

While the Court has shied away from strictly categorizing guarantees as ‘rights’ or ‘freedoms’, there can be little doubt that the questions “can the state prevent me from building a church?” or “can the government criminalize my political message?” are categorically different than “must the government purchase a parcel of land for my church?” or “must the legislature force private broadcasters to disseminate message?” While both state action and state inaction can operate to effectively ‘preclude’ the meaningful exercise of one’s substantive freedom, depending on the circumstances, the two inquiries are and must be treated differently as a matter of constitutional law. Simply stating that the line between ‘rights’ and ‘freedoms’ can occasionally be a hazy one cannot obliterate the line entirely.

On a more practical level, a ‘right to the meaningful exercise of a freedom’ standard necessarily requires the courts to attempt the almost impossible task of determining with any degree of certainty what is required to ‘meaningfully exercise’ a freedom – at what degree of meaningfulness does the state obligation to enhance the purposes and objectives of the association, expression or religion, begin and end? Many freedoms that the state may not unjustifiably encumber, such as writing papers on constitutional interpretation in obscure legal journals, may be done in vain, but that does not normally entitle authors to constitutional remedy.

Benjamin Oliphant, “Exiting the Freedom of Association Labyrinth:  Resurrecting the Parallel Liberty Standard Under 2(d) & Saving the Freedom to Strike” (2012), 70 UTFLR 36 at 68-71.

In other words, I think there is a fundamental distinction between section 15, relative – entitlements – you cannot deprive me of this benefit you give to everyone else – and the task of defining baseline entitlements to ‘meaningful freedom’, and we should not conflate them.  It is one thing for the Courts to erect a wall of freedom over which the state may not intrude, but quite another to start directing the distribution of government funds and legislative protections to achieve some abstract amount of ‘freedom’. The latter strikes me as a rather massive expansion of judicial power, made deceptively easy in the labour relations context by a ready-made statutory superstructure of positive protections (i.e. the Wagner Act labour relations legislation established across the country).

The Counter Point

All of that being said, I agree that the distinctions accepted in a range of cases discussed above – between a freedom and a right, between positive and negative entitlements, between state action and inaction, and so on – are not bright line rules that can resolve hard constitutional cases on the basis of an initial characterization.

To see how the distinction can break down, consider the Government conditioning access to a public space – e.g. holding a political rally in a city park – on supporting a particular partisan viewpoint.  I should think that a rather intolerable intrusion upon freedom of expression, despite the fact that a publicly-maintained public park could be characterized as access to a ‘platform’ of sorts. So the mere fact that a claim can be characterized as access to a state ‘platform’ or ‘benefit’ cannot end the analysis.

The US case law here might be instructive. The general rule applied by the US Courts is consistent with the orthodox position stated above. As recently outlined in AID v. Alliance for Open Society Intern.:

As a general matter, if a party objects to a condition on the receipt of federal funding, its recourse is to decline the funds. This remains true when the objection is that a condition may affect the recipient’s exercise of its First Amendment rights. See, e.g., United States v. American Library Assn., Inc., 539 U.S. 194, 212, 123 S.Ct. 2297, 156 L.Ed.2d 221 (2003) (plurality opinion) (rejecting a claim by public libraries that conditioning funds for Internet access on the libraries’ installing filtering software violated their First Amendment rights, explaining that “[t]o the extent that libraries wish to offer unfiltered access, they are free to do so without federal assistance”); Regan v. Taxation With Representation of Wash., 461 U.S. 540, 546, 103 S.Ct. 1997, 76 L.Ed.2d 129 (1983) (dismissing “the notion that First Amendment rights are somehow not fully realized unless they are subsidized by the State” (internal quotation marks omitted)). [at 2328]

One of the cases cited for the general rule, Regan v. Taxation With Representation of Wash, closely resembles the CWP challenge, as it deals with restrictions on charitable tax status. There the Court rejected the premise that freedom of expression requires the Government to extend state subsidies for public activities, stating:

The reasoning of these decisions is simple: “although government may not place obstacles in the path of a [person’s] exercise of . . . freedom of [speech], it need not remove those not of its own creation.” Harris, 448 U. S., at 316. Although TWR does not have as much money as it wants, and thus cannot exercise its freedom of speech as much as it would like, the Constitution “does not confer an entitlement to such funds as may be necessary to realize all the advantages of that freedom.” Id., at 318. (…)

TWR contends that § 501(c)(3) organizations could better advance their charitable purposes if they were permitted to engage in substantial lobbying. This may well be true. But Congress — not TWR or this Court — has the authority to determine whether the advantage the public would receive from additional lobbying by charities is worth the money the public would pay to subsidize that lobbying, and other disadvantages that might accompany that lobbying. (…) It is not irrational for Congress to decide that tax-exempt charities such as TWR should not further benefit at the expense of taxpayers at large by obtaining a further subsidy for lobbying. [at 549; emphasis added]

As in Canada, however, this is not a hard and fast rule. I don’t presume to be an expert in First Amendment jurisprudence, but it seems that they have come up with a few subsidiary doctrines to deal with borderline cases where an otherwise meritorious claim could be characterized as one to state support.

First, if the ‘benefit’ being denied is one traditionally available to all to use for the purposes of expression (as in our public parks example above), the government cannot deny it without a good reason: “the existence of a Government “subsidy,” in the form of Government-owned property, does not justify the restriction of speech in areas that have “been traditionally open to the public for expressive activity” (see United States v. Kokinda, 497 U. S. 720, 726 (1990)).

Second, in the AID case quoted above, the majority found that the Government could not make federal funding for a program contingent on engaging in expressive activities that are in some sense beyond the scope of the objectives of the program itself, and perhaps especially where the condition “requir[es] recipients to profess a specific belief”. This distinction is itself a fine one, and I think that particular case shades into a different area – the coerced expression cases. However, because the CWP case does not involve requiring the endorsement or forswearing of any particular political message, I am not sure this angle will be helpful.

Third, and most relevant to the CWP challenge, the US Supreme Court seems to be relatively comfortable with restricting the use of public funds and subsidies for certain expressive activities, but much less comfortable with blanket prohibitions tied to the group itself. As explained in Rust v. Sullivan:

In contrast, our “unconstitutional conditions” cases involve situations in which the Government has placed a condition on the recipient of the subsidy rather than on a particular program or service, thus effectively prohibiting the recipient from engaging in the protected conduct outside the scope of the federally funded program. In FCC v. League of Women Voters of Cal., we invalidated a federal law providing that noncommercial television and radio stations that receive federal grants may not “engage in editorializing.” Under that law, a recipient of federal funds was “barred absolutely from all editorializing” because it “is not able to segregate its activities according to the source of its funding” and thus “has no way of limiting the use of its federal funds to all noneditorializing activities.” The effect of the law was that “a noncommercial educational station that receives only 1% of its overall income from [federal] grants is barred absolutely from all editorializing” and “barred from using even wholly private funds to finance its editorial activity.” 468 U. S., at 400. We expressly recognized, however, that were Congress to permit the recipient stations to “establish ‘affiliate’ organizations which could then use the station’s facilities to editorialize with nonfederal funds, such a statutory mechanism would plainly be valid.” Ibid. Such a scheme would permit the station “to make known its views on matters of public importance through its nonfederally funded, editorializing affiliate without losing federal grants for its noneditorializing broadcast activities.” Ibid.

Similarly, in Regan we held that Congress could, in the exercise of its spending power, reasonably refuse to subsidize the lobbying activities of tax-exempt charitable organizations by prohibiting such organizations from using tax-deductible contributions to support their lobbying efforts. In so holding, we explained that such organizations remained free “to receive deductible contributions to support … nonlobbying activit[ies].” 461 U. S., at 545. Thus, a charitable organization could create, under § 501(c)(3) of the Internal Revenue Code of 1954, 26 U. S. C. § 501(c)(3), an affiliate to conduct its nonlobbying activities using tax-deductible contributions, and at the same time establish, under § 501 (c)(4), a separate affiliate to pursue its lobbying efforts without such contributions. 461 U. S., at 544. Given that alternative, the Court concluded that “Congress has not infringed any First Amendment rights or regulated any First Amendment activity[; it] has simply chosen not to pay for [appellee’s] lobbying.” Id., at 546. We also noted that appellee “would, of course, have to ensure that the § 501(c)(3) organization did not subsidize the § 501(c)(4) organization; otherwise, public funds might be spent on an activity Congress chose not to subsidize.” Id., at 544. The condition that federal funds will be used only to further the purposes of a grant does not violate constitutional rights. “Congress could, for example, grant funds to an organization dedicated to combating teenage drug abuse, but condition the grant by providing that none of the money received from Congress should be used to lobby state legislatures.” See id., at 548. [at 197-198; emphasis added]

If I were a betting man, I would suspect the Canadian courts might find this compromise position appealing. That is, while the government could condition the receipt of funds (or tax breaks) on using that money for certain purposes and not others (i.e. political activities), it cannot then prevent the organization from using funds collected from other sources for those political purposes. Whether this is a sustainable distinction in principle or a workable idea in practice is something I will leave to others.

Conclusion

The challenge raised by CWP is not an easy one to resolve.  However, it is a good one to test the theory of labour union exceptionalism in the context of section 2, i.e., that the courts are willing to extend relatively robust “positive” protections to labour unions to make their freedom of association more meaningful, in a way they would typically not contemplate in other contexts, either for other associations who might benefit from additional statutory protections (as most associations would), or for other persons who could benefit from government ‘enhanced’ freedom of expression or religion.

I should clarify that I do not mean to suggest that there’s any sort of ideological predisposition in favour of labour unions, at the Supreme Court level or otherwise. I suspect the discrepancy in the case law, if there is one, is better explained by path dependency and a sense of fairness than either some high constitutional principle or bias. That is, because the Wagner Act model has been extending affirmative rights to labour unions for nearly a century, and these rights were in exchange for a rather dramatic diminution of the freedom of workers (discussed here at 260 n. 28), the Court appears uncomfortable with legislation that fails to extend its baseline protections, even if there would be no independent constitutional entitlement to these particular statutory protections but for the historical fact of the Wagner Act model.

Nevertheless, in light of the relative success labour unions have had in claiming positive protections under freedom of association as compared with others in the context of the other fundamental freedoms, we should not be surprised that the CWP included a freedom of association claim in their Application, using language that conspicuously mirrors the that the Court has used in the context of extending affirmative protections to labour unions (see Notice of Application at paras 23-27).

If nothing else, this will put the courts to the task of deciding whether there are other associations beyond labour unions that require positive state protection or support to make their expressive and associational activities sufficiently “meaningful” to pass constitutional muster. This becomes a hard question once we realize that every organizations “freedom” to achieve their objectives and purposes would be enhanced in so far as they received government funding or positive statutory protection not available to everyone else.

At the very least, there is no doubt that the CWP’s expressive and associational activities will be enhanced to the extent that it would continue to benefit from beneficial tax treatment, without the corresponding responsibility to refrain from engaging in “political activities”. The question is whether the government is constitutionally obliged to support the expressive and associational activities that help the CWP advance its mission. Or, put differently, is there some principled reason why CWP’s claim to ‘meaningful’ expression and association does not require affirmative state protection of this sort, but that labour unions are entitled to certain positive protections to enhance the meaningful exercise of their freedoms? I think the CWP deserves a good answer to this question.

 

* First, the idea that the impugned ITA provision has an unconstitutional purpose , which the CWP’s Notice of Application asserts, might be a clever end run around all of the above, and raises other complications that I have not addressed in this post. For my purposes, I have assumed that there is some sort of rational basis for the provision that extends beyond the mere objective of repressing political expression, as such. Second, I recall there being allegations that the audits being undertaken by the previous government were politically motivated. I have no idea whether this is true and this does not appear to form the basis of CWP’s Application, and so I have assumed that not to be the case for the purposes of the post. If that were the essence of the allegation, however, it would raise constitutional issues, whether or not the impugned provision is permissible as a general rule.  In particular, I think that would get us into whether the government had an unconstitutional purpose in deciding to audit particular organizations, which might involve a Doré type analysis (scrutinizing the administrative discretion exercised by the CRA), or perhaps a Little Sisters type challenge, both of which raise complications I have not addressed.

 

Constitutional Purposes vs. Constitutional Text: On R. v. Pino

In my previous guest post at Double Aspect, I asked an intractable question: what is it that we are doing when we are engaged in constitutional interpretation?  Depending on how one answers this question, different sources of meaning will become more or less significant.  However, one source must always be at least relevant: the Constitutional text itself.  It is the one, and perhaps only one, fixed star in the constellation of sources one draws upon in the course of constitutional interpretation, and I have previously argued that we should pay particular attention to it as a result.  The judicial conscience or temperament, social values, public policy objectives, academic prescriptions, international law, and many other potential sources, may differ between reasonable people and may change with the times. The actual Constitutional text can and will not, and it is only source that has withstood the crucible of democratic decision-making and been enacted into law. While the text alone cannot answer many of our difficult questions, when it does give an answer – or, as importantly, rule out an answer – we should consider that significant.

Of course, no plausible approach to constitutional interpretation self-consciously ignores the text or treats it as irrelevant– but some pay greater attention to it than others.  A recent Ontario Court of Appeal decision, R. v. Pino, asks us to consider how much attention is enough.

I will not get too deep into the facts of that case, because they tend to complicate matters, and I happen to think that the very same outcome was available by another route without getting into the interpretive swamp into which I now leap. But in very brief: the police received a tip that the accused was running a grow-op, which seemed to be borne out by subsequent observation of her residence; the police followed the suspect, obtained incriminating evidence from the trunk of her car in a lawful search incident to arrest, but committed Charter violations in the course of that arrest, then failed to adequately inform the accused of her right to counsel, and denied that right for a period of time later on.

The issue I am interested in was the Court’s conclusion that evidence otherwise collected lawfully can be considered “obtained in a manner” that infringed the Charter, due to a subsequent Charter breach. The section at issue is section 24(2), which provides:

(2) Where, in proceedings under subsection (1), a court concludes that evidence was obtained in a manner that infringed or denied any rights or freedoms guaranteed by this Charter, the evidence shall be excluded if it is established that, having regard to all the circumstances, the admission of it in the proceedings would bring the administration of justice into disrepute. [emphasis added]

As the Court of Appeal in Pino notes, this provision, plainly read, seem to require a causal connection between the Charter breach and the obtaining of the evidence:

On a superficial reading of s. 24(2) one might be tempted to conclude that the “obtained in a manner” requirement can only be met by a causal connection between the breach and the discovery of the evidence: “but for” the breach the evidence would not have been discovered. [para 50]

However, the Court found that one should resist that temptation, because “the Supreme Court has long recognized that a causal connection is unnecessary” [para 50]. We should pause to consider why.

As I read the Supreme Court’s decision in R. v. Strachan (and I confess that I am slightly outside my comfort zone here), the primary reason the Court gives for why “obtained in a manner” does not require a causal connection between a breach and the discovery of evidence is because it should not. The Court says that including a causation requirement (an unavoidable and familiar inquiry in many areas of law, but leave that aside) would present a “host of difficulties” [at para 39]; it would be ungenerous and restrictive [at para 40-42]; and it would prevent the courts from reaching what the Court considers to be the “more important” branch of the 24(2) inquiry, which is whether the admission of the evidence would bring the administration of justice into disrepute [at para 47].  In short, requiring a causal connection would be not “useful”, “fruitful”, or ‘sensible’; a “better approach”, according to the Court, “would be to consider all evidence gathered following a violation of a Charter right, including the right to counsel, as within the scope of s. 24(2)” [para 45].

These may all be excellent reasons not to link the exclusion of evidence to whether or not they were “obtained in a manner” that breached the Charter. But as a matter of interpreting that language, the logic (taken alone) is somewhat unsatisfying: just because one interpretation would have consequences an interpreter does not consider sensible does not necessarily confer a license to circumvent an authoritative direction. (The fact that it may not be sensible to locate a stop sign on a normally-vacant rural road does mean that the sign actually says “yield”.)

The Court of Appeal in Pino follows the Strachan logic quite closely, and expands it quite naturally, to capture Charter breaches that occurred after the evidence was obtained. In coming to that conclusion, the Court quoted the eminent Professor Kent Roach, who points out that:

From a regulatory perspective, it should not matter whether the evidence was obtained before or after a serious Charter violation. In both cases, the administration of justice could be brought into disrepute if the courts appear to condone a serious Charter violation. If the court is concerned with responding to serious violations, there is no reason why evidence discovered before a violation should not be considered for exclusion. [para 69]

Thus, and notwithstanding the Supreme Court’s apparent direction that evidence must be gathered following a breach to qualify for exclusion under 24(2) [Pino at para 63], the Court of Appeal devised an approach it considered more sensible, and better directed at achieving the purpose of the section:

A generous approach to the “obtained in a manner” requirement makes good sense because this requirement is just the gateway to the focus of s. 24(2) – whether the admission of the evidence would bring the administration of justice into disrepute. (…)

So, should it make a difference whether the s. 10(b) breaches occurred before or after the discovery of the evidence? I do not think so. In either case, the administration of justice could be brought into disrepute if the court condoned serious Charter violations. [paras 56, 77, emphasis added]

In short, the Court appears to reason that the “overall purpose of the section”, which it finds is to determine “whether admission of the evidence would bring the administration of justice into disrepute” [at para 51], would not be achieved by being too finicky about exactly how or when that evidence itself was obtained.

However, as the Court of Appeal itself recognized earlier in the judgment, 24(2) imposes two requirements for the exclusion of evidence: 1) that it was obtained in a manner that violates the Charter; and 2) that its admission would bring the administration of justice into disrepute (see Strachan at para 36, Pino at paras 35-36). A provision permitting the exclusion of evidence wherever the administration of justice would be brought into disrepute could be easily drafted. Here, look, I’ll draft one now:

(2) If a person charged with an offence has had their Charter rights breached, the Court may exclude any evidence if the admission of it in the proceedings would bring the administration of justice into disrepute.

But that is not the provision in the Charter, and I think we should be cautious before reading it in that manner.

In fairness, the Court of Appeal did not go that far, emphasizing that there must at least be some “contextual” link between the breach and the evidence, whatever the sequencing of the two [paras 72-74].  Returning to its reasoning, the Court of Appeal quoted with approval the trial judge’s extra-judicial writings, in which he had observed:

There are also sound reasons of policy for leaving this door open. Assume that the police discover marijuana during a lawful and reasonable pat-down search and then publicly and needlessly go on to strip search the suspect. Is a court to be deprived of the power to exclude the evidence because of the sequence of events? To insist on the breach preceding the discovery of evidence as an absolute precondition to exclusion means that ex hypothesi evidence can be admitted even where its admission would bring the administration of justice into disrepute, just because of the order in which things happened to occur.

As in Strachan, this is an excellent reason for not including certain words (like “obtained in a manner”) in a constitutional provision, but with great respect, I am not sure how far the logic takes us as a matter of interpreting what those words mean.  One does not typically ignore one requirement because applying it would prevent him or her from considering the second requirement. That is not normally how “requirements” work. (It might be how “gateways” work, to use the Court of Appeal’s word, but only to the extent that gateways not actually require anything or serve any discernable purpose.)

This is a relatively intuitive point that I fear can be obscured by lawyerly creativity, so consider a direction from your parents: “If you want dessert, you may have some, but only if you finish your supper first”. No non-lawyer familiar with the English language would suggest that a child may ignore the supper requirement because it would be impractical in some circumstances to eat supper, or because it would prevent the child from focusing on the “more important” inquiry of whether the child wants to have dessert.

And I am not sure that this trouble can be avoided by reciting the words “broad” and “generous” over and over. It is worth pointing out that every time evidence is excluded from a proceeding, society’s interest in having persons held accountable for the commission of offences is compromised. Accordingly, the old common law rule was that all relevant evidence was admissible, with courts adverse to the conclusion that the “criminal is to go free because the constable has blundered”. Section 24(2) has unequivocally changed that, and to many of us, for the better; society, like the accused, also have an interest in law enforcement acting in a constitutionally sound manner. But that does not make the exclusion of evidence somehow costless, and on some level, it must have been that trade-off motivating those charged with drafting and approving the “obtained in a manner” requirement.

The same tension exists whenever it is the constitutionality of a statute that is at issue. Every time the Court reads a Charter right or freedom broadly and generously, they necessarily read the corresponding scope of democratic governance narrowly and restrictively.  Of course, this tension is an unavoidable (and expected) result of enacting constitutional rights; every application of the Charter, no matter how narrowly interpreted, limits the scope of democratic self-governance to some extent, which is entirely the point of entrenching constitutional rights.  So pointing out consequences is not suggest the courts should ignore their constitutional function; it is only to suggest that any gain in terms of rights comes at a cost in terms of democracy.  As such, an overly broad interpretation results in an unduly narrow scope for democratic self-governance, and vice versa.  It is not obvious to me that either error should be deemed presumptively preferable, and the proposition that Charter rights should necessarily and always be interpreted as broadly as possible is more often asserted than defended.  A sound interpretation of any Charter provision, including section 24(2), may be broad, or it may be narrow, or it may be in between, but presumably that should be the conclusion of the interpretive process, not the predetermined objective.

One other point might be made. With each step we take away from genuinely trying to understand what the text means, the next step further away gets easier. The Court in Strachan found that the phrase “obtained in a manner” imports no strict causal requirement, but nevertheless stated that any breach at least had to precede the discovery of the evidence. This chronological threshold has now been scrapped, essentially because it made no sense once you have deemed causation to be irrelevant. The next judicial step may be further away, still: if maintaining the repute of the administration of justice is our primary (or exclusive?) goal, why should we insist on any meaningful connection between the breach and the discovery of evidence? Is that requirement not ‘artificial’, in the same way as the other thresholds; i.e. because, in certain cases, it may prevent us from reaching the “more important” inquiry of whether admission of the evidence would bring the administration of justice into disrepute? For instance, why should evidence collected lawfully be excluded because a suspect was beaten savagely in the course of an immediately succeeding arrest, but not if the savage beating occurred back at the police station a few hours, a day, or two days later?

The point is that while each step away from the one before may seem minor and defensible enough, if we keep walking we may eventually lose sight of where we started. I should state my view that the Court of Appeal’s decision in Pino has much to recommend it: while a chronological requirement makes sense with a causation requirement (something cannot be caused by something that comes after it), it does not seem to serve any obvious purpose if we are not at all concerned with causation. Thus, as a matter of reading the case law, I’m not sure the Court of Appeal’s decision can be gainsaid. But as a matter of reading the Constitution, I wonder if we have not gone too far, or are at least teetering on the edge: we have now arrived at a situation where evidence may be considered to have been “obtained in a manner” that breaches a Charter right, because a Charter right was breached after the evidence was “obtained in a manner” that does not breach a Charter right.

Professor Sankoff made something like this point on twitter, asking: “isn’t it just time to stop pretending that ‘obtained in a manner’ has any meaning?” Now, I do not take the Professor to be implying that the words are actually meaningless; “obtained in a manner” is not gibberish, it is a grammatically sound and intelligible English phrase. The question he may have been asking is whether we should stop pretending that we care what that meaning is, and that is a fair question, if the answer could possibly be “no”.

Finally, and I cannot stress this next point enough: downplaying the ‘obtained in a manner’ requirement might result in a better provision; it may be easier to apply; it may be a more just and sensible provision; and I may well prefer it myself, if I had my druthers. So I do not question whether where we have ended or might ultimately end up may be preferable in many respects; I question whether implicitly redrafting the language in a way that is more preferable constitutes an “interpretation”, or something else.

As I’ve argued before, where judicially-defined purposes not only inform but overwhelm the text, we might well end up with a better constitution; but it will be one to which no democratically elected body assented. Maybe we don’t care about that, or at least not very much, as long as we approve of the results. And we can certainly take solace in the fact that our judiciary is comprised of thoughtful, principled, and eminently – well, sensible – people, and I personally have no doubt on that score.  If I were to choose a group of people I trust with writing a new constitution, our judges would be among the top of my list.  But for the time being, they have been tasked with interpreting the Constitution written and approved by others, not to write it anew. While the line between the two functions is murky, there must be a line.

Bilingualism, the SCA Reference, and Buffet-Line Constitutional Interpretation

Professors Grammond and Glover, as well as my gracious host Léonid Sirota, have all addressed the constitutionality of requiring judges to be bilingual in order to be qualified for appointment to the Supreme Court. In my view, all are excellent efforts to come to grip with difficult constitutional problems, and taken alone, I find each of them plausible and compelling.  Taken together, they reveal a larger maxim: that we cannot agree upon an answer to something unless we agree what the question is. That is: what are we looking for when we are doing constitutional interpretation? Are we looking for the framers’ specific intentions, whether or not those intentions are successfully manifested in constitutional language? Should we seek to achieve the framers general purposes, derived from any range of sources? Are we looking for the meaning of the words that they set out, as understood by a reasonable person? Are we seeking an interpretation that best fits with the structure of our institutions? Or what we need as a society (distilled through the views and opinions of nine eminent jurists) at the present moment?

Each of these options is reflected in the approaches offered so far, and unsurprisingly, they lead to very different outcomes, and different readings of the SCA Reference. I provide my own tentative take on each of the arguments presented so far, a task I undertake with some trepidation, as I’m convinced I have not thought as long, or read as widely, about this specific issue as these authors have. So I invite them (and others) to issue corrections or rebuttals in the comments.

I will begin with Professor Grammond’s piece. As I read it, his approach identifies two limits on the power of Parliament to unilaterally alter the composition of the Court, and each are grounded primarily in different methods of constitutional interpretation. First, he suggests that Parliament cannot unilaterally alter “Quebec’s representation on the Court”, a restriction he finds grounded in a principle that is (as Sirota noted) thoroughly originalist: Quebec’s representation is considered amongst the “specific issues in respect of which there is cogent evidence that the framers of the 1982 Constitution intended to freeze the status quo”.

Prof. Grammond’s second limit is not originalist, as such, but what we might call “structuralist”: Parliament cannot unilaterally alter the court’s composition in a manner that undermines “the Court’s role as a final court of appeal and the Court’s independence”, or that otherwise effects truly fundamental changes to the institutional functioning of the Court. As I understand it, the problem with such changes is not necessarily that they defy any clear proscription in the language of the constitution, but that they undermine the core structure of the institution those provisions create.

As making bilingualism a condition of eligibility would offend neither of these limits, Prof. Grammond concludes that no constitutional amendment is required for that proposed change. He notes that this interpretation seems to run afoul of the Supreme Court’s ostensibly specific direction that any substantive changes to the eligibility requirements found in ss. 4-6 of the SCA constitute an amendment. In response, Prof. Grammond cautions us not to be too persnickety about each and every word or phrase in the SCA Reference; the judgment should not be read like a statute, he suggests, but rather in a more flexible, common law fashion.

I do not necessarily take issue with that, as a general proposition. My difficulty with Prof. Grammond’s argument is less that it requires some artful dodging around this or that particular clause in the judgment, but that it does not seem that the limits he proposes actually justify the holding in the SCA Reference. I think Prof. Grammond comes close to acknowledging this, in noting that the logic underlying that judgment did not actually explain “why someone in the position of Justice Nadon could not be said to represent Quebec adequately”.  I will return to this point in a moment.

Mr. Sirota takes a different interpretive tack, which I will describe as more “textualist” in nature. He criticizes Prof. Grammond’s use of the framers’ intentions as the appropriate constitutional lodestar, and failing to identify why some eligibility requirements alter the Court’s “composition” while others do not. It is not obvious to me from his short piece what Sirota himself thinks the word “composition” means, exactly, however based on the holding of the SCA Reference, he reasonably concludes that the Court has already determined what the word “composition” means, and it means those eligibility requirements set out in ss. 4-6 of the SCA.  As I understand the position, if composition includes eligibility requirements, as the Court plainly stated, then adding an eligibility requirement would require an amendment, just as much as removing one.

In a tightly argued response, Prof. Glover implies that Sirota, at least, is not asking the right question: the right question is not whether the precise form of those sections would change, but whether the proposed change is one of substance.  To Prof. Glover, the best reading of the logic of the SCA Reference is that it shields from unilateral change “only those aspects of composition and eligibility that are necessary to ensure the Court’s competence, legitimacy, integrity, and proper role and functioning”.

It seems to me that this approach is not originalist, or textualist, but some form of purposivm: what were the provisions designed to achieve, and what types of restrictions are necessary to achieve those ends? As their purpose is to ensure the Court’s competence, legitimacy, integrity, role and functioning, it is against these ends that we measure whether a unilateral amendment would be permitted. Prof. Glover finds that it would not, as a requirement of bilingualism would be a change linked to the “judicial competency and institutional integrity of the Court”.

As I see it, Prof. Glover’s dispute with Sirota is relatively fundamental: should the Court be looking to give the term “composition” a fixed and steady meaning consistent with past precedent, or looking at what objectives entrenching the Court’s “composition” were intended to achieve? By contrast, Prof Glover’s dispute with Prof. Grammond is less stark.  While for Prof. Grammond, the core purposes of entrenching the composition of the court are to protect the fundamental nature of the institution, including Quebec’s representation and the Court’s role as an independent, final court of appeal, Prof. Glover sees the purposes as slightly more demanding.

Truth be told, Prof. Glover’s analysis reads closest to the way I would expect the Court to approach the issue. I would however argue that Prof. Glover’s position is vulnerable to the same criticism as Prof. Grammond’s: it does not, in my view, adequately explain the holding of SCA Reference itself. Again, Prof. Glover’s constitutional analysis seems to hinge on what she views as the purpose of the provisions, not the specific way that those purposes have been put into practice (for instance, the precise criteria in ss.4-6 of the SCA). It is for this reason that she can argue that less substantive changes, such as to allow the appointment of advocates of at least 9 years standing at the bar of a province (instead of 10), would not require an amendment, although it would clearly require a change to ss. 4-6 of the SCA.

However, I think that in order for the SCA Reference to be correctly decided, the specific criteria in the SCA are critical, not merely the purposes underlying them. Indeed, the Court’s logic suggests that it did not matter whether the means chosen (i.e. the specific criteria) by Parliament would best achieve its purpose, or whether other criteria would meet them as well or better.  What matters was what criteria Parliament chose:

It might be argued that excluding former advocates of at least 10 years standing at the Quebec bar does not perfectly advance this twofold purpose because it might exclude from appointment candidates who have civil law expertise and who would in fact bring Quebec’s legal traditions and social values to the Court. In other words, it could be argued that our reading of s. 6 is under-inclusive when measured against the provision’s objectives.

This argument is not convincing. Parliament could have adopted different criteria to achieve the twofold objectives of s. 6 — for instance by requiring a qualitative assessment of a candidate’s expertise in Quebec’s civil law and legal traditions — but instead it chose to advance the provision’s objectives by specifying objective criteria for appointment to one of the Quebec seats on the Court. In the final analysis, lawmakers must draw lines. The criteria chosen by Parliament might not achieve perfection, but they do serve to advance the provision’s purpose: see Michael Plaxton and Carissima Mathen, “Purposive Interpretation, Quebec, and the Supreme Court Act” (2013), 22 Const. Forum 15, at pp. 20-22. [paras 57-58, emphasis added]

While this passage is found in the Court’s interpretation of ss. 5 and 6, the Court went on to find that the precise outcome of that interpretation was constitutionally entrenched, such that a rather minor change to the language of those provisions – so minor as to accord with what many plausibly thought those provisions meant in the first place – would thereby constitute a constitutional amendment.

Thus, if our guideposts are not the precise language and exact criteria in ss. 4-6 of the SCA, but rather the more broadly understood purposes behind them (such as “the requirement that Quebec be meaningfully represented on the Court”), then it is difficult to see how a change permitting the appointment of, say, a Quebec-born and raised francophone, who had 20 years’ experience at the Quebec bar and considerable experience in civil law, would substantively undermine meaningful representation for Quebec.  What it did, according to the Supreme Court, was something different: it offended the “specific eligibility requirements for appointment from Quebec” (para 105), which, I think, the Court equated with those specifically set out in the SCA.

Notwithstanding these reservations, I think there is much support in the SCA Reference for all of the positions put forward, particularly once we take Prof. Grammond’s caution and realize that the Court is not likely to consider itself forever hidebound by the precise holding in that decision. Indeed, if we take that caution and simply rely on accepted methods of constitutional interpretation, the odds are even more clearly divided.  All three of the arguments begin with entirely sound premises (one of many widely-accepted methods of interpretation), and follow them logically to perfectly defensible conclusions. That is to say, each argument is credible and compelling in its own right, notwithstanding that all three authors are looking for very different things and arrive at fundamentally different conclusions.

Perhaps this degree of quantum uncertainty is unavoidable in hard case like this. But I cannot help but think that if this is a problem, it is exacerbated by the Court’s jurisprudence, which sometimes resembles a buffet-line approach to constitutional interpretation, authorizing interpreters to pick or choose what method of interpretation they will follow and in what measure, and without any guidelines as to how to discard, reconcile, combine, or choose between the methods on offer. As judges are often fully liberated to determine what the question is, it may be hard to prevent them from arriving at their preferred answer.