Should Homelessness be an Analogous Ground? Clarifying the Multi-Variable Approach to Section 15 of the Charter

By: Joshua Sealy-Harrington

PDF Version: Should Homelessness be an Analogous Ground? Clarifying the Multi-Variable Approach to Section 15 of the Charter 

Case Commented On: Tanudjaja v Canada (Attorney General), 2013 ONSC 5410

This post discusses a decision from the Ontario Superior Court which rejected “homelessness” as an analogous ground under section 15 of the Canadian Charter of Rights and Freedoms. I divide the analysis into two sections: (1) a discussion of the analytical flaws in the court’s approach to analogous grounds, and (2) an application of a multi-variable approach to the potential ground of homelessness to demonstrate its apparent viability as an analogous ground. This case should be of concern to ABlawg readers interested in section 15 jurisprudence as well as those advocating for the interests of the homeless.                                                                   

(1) A Flawed Approach to Homelessness as an Analogous Ground

In Tanudjaja v Canada (Attorney General), 2011 ONSC 5410, Lederer J of the Ontario Superior Court of Justice allowed motions by the Federal Crown and the Ontario Provincial Crown for an order dismissing the application by Tanudjaja and others for Charter relief. In particular, Lederer J allowed motions dismissing applications that alleged breaches of sections 7 and 15 of the Charter flowing from legislative changes that resulted in increased homelessness and inadequate housing.

As a preliminary note, Lederer J’s analysis of homelessness as an analogous ground is obiter dicta. He concludes that whether or not homelessness is an analogous ground is irrelevant because “the actions and decisions complained of do not deny the homeless a benefit Canada and Ontario provide to others or impose a burden not levied on others” (at para 128). Despite its obiter nature, my discussion focuses exclusively on his analysis regarding analogous grounds.

Lederer J alludes to the various factors that have been considered in the jurisprudence for assessing analogous grounds (at para 127). However, he ultimately rejects homelessness as an analogous ground because of a novel requirement for analogous grounds that he identifies and which I will label as “definability” (at para 134). In his own words, a lack of definability may be described as follows:

[131] [T]here is no means to understand the parameters that would define those who make up the analogous group. Who would be the members? On what basis is the group said to be analogous? In these circumstances, it is impossible to come to a substantive understanding of what the analogous ground is.

To be clear, according to Lederer J, an indefinable ground is one which fails to outline an objectively ascertainable group of people that belong to it. As he explains with respect to homelessness:

[129] In the circumstances of this Application, it is not possible to identify who is “homeless” […] homelessness is not, for the purposes of this Application, restricted to those without homes. […] It may be that what is being referred to as “the homeless” includes those without “affordable, adequate and accessible” housing. What is adequate housing? Presumably, this depends on the circumstances of the individuals involved.

I have three observations about Lederer J’s approach to analogous grounds: (a) he lacks a jurisprudential basis for a “definability” requirement, (b) he appears to conflate definability with a different idea (heterogeneity) in his reasoning which results in a lack of analytical clarity, and (c) neither definability nor heterogeneity are conclusive when assessing analogous grounds.

(a) Why Definability is Not a Requirement of the Analogous Grounds in the Jurisprudence

In support of a purported definability requirement for analogous grounds, Lederer J cites Polewsky v Home Hardware Stores Ltd (1999), 68 CRR (2d) 330 (ONSC) [Polewsky], where “poverty” was rejected as an analogous ground. He also claims that the Ontario Court of Appeal “recognized this limitation” (that is, the definability requirement) in Falkiner v Ontario (Ministry of Community and Social Services) (2002), 59 OR (3d) 481 (CA) [Falkiner], where “social assistance recipients” was recognized as an analogous ground.

With regard to Falkiner, the Court of Appeal did not adopt a definability requirement in its reasoning. Rather, in that case, social assistance recipients was recognized as an analogous ground after a consideration of many other factors including historical disadvantage and continuing prejudice, immutability, protection in human rights statutes, and heterogeneity (at paras 82-93).

The other key decision cited by Lederer J – Polewsky – in one passage appears to consider a definability requirement:

[59] A third reason lies in a consideration of those who make up the group of people who are in financial need. The poor in Canadian society are not a group in which the members are linked by shared personal or group characteristics. The absence of common or shared characteristics means, in my view, that poverty is not an analogous grounds to those enumerated. Those enumerated grounds are defined by one or more shared characteristics whether it be race, nationality, colour, religion, sex, age or disability.

Arguably, lacking a “shared characteristic” could be interpreted as lacking definition (in so far as groups may be defined by their shared characteristics). However, upon further inspection, that is not the interpretation advanced by Gillese J in Polewsky.

Later in her reasoning, Gillese J clarifies that lacking a shared personal characteristic is equivalent with being a “disparate and heterogeneous group” (at para 60). In particular, Gillese J provides the example of agricultural workers as such a disparate and heterogeneous group. However, agricultural workers are also definable (by a common occupation), which in turn precludes the possibility that the court was applying a definability requirement. As a consequence, Polewsky, like the other authorities that Lederer J relies on, cannot be read as creating a definability requirement.

(b) Conflating Indefinability with Heterogeneity

For the sake of clarity, I will begin with a basic definition of indefinable and heterogeneous groups. An indefinable group is one which lacks clear parameters for identifying its members. A heterogeneous group is one which has a diverse membership.

Lederer J’s reliance on Polewsky as a key authority results in unclear reasoning. In particular, Lederer J seems to conflate his basis for rejecting homelessness as an analogous ground (indefinability) with a factor considered in Polewsky when rejecting poverty as an analogous ground(heterogeneity).

This conflation of concepts is demonstrated in Lederer J’s closing remarks, where his basis for rejecting homelessness as an analogous ground subtly transitions from indefinability to heterogeneity. First, he rejects homelessness as an analogous ground because it is indefinable:

[134] Homeless is not a term that, in the context of this case, can be understood. Without an understanding of the common characteristics which defines the group, it cannot be established as an analogous ground under s. 15(1) of the Charter (emphasis added).

Then, he discusses the heterogeneity of homelessness:

[135] There is a list of groups which are said, in the Application, to be protected from discrimination under s. 15(1) of the Charter and disproportionately affected by the lack of adequate housing. It includes: “women, single mothers, persons with mental and physical disabilities, Aboriginal persons, seniors, youth, racialized persons, newcomers and persons in receipt of social assistance” […] What discrimination can there be when all of the groups identified as being subject to this discrimination, taken together, include virtually all of us? (emphasis added).

Finally, he again rejects homelessness as an analogous ground, this time because it is heterogeneous:

[136] Homelessness is not an analogous ground under s. 15(1) of the Charter. The Application does not propose to protect “discreet and insular minorities”. It is an attempt to take “disparate and heterogeneous groups” and treat them as an analogous ground under s. 15(1) of the Charter. Such groups do not obtain this protection (emphasis added).

However, “heterogeneous” groups are not necessarily “indefinable.” In fact, broad and diverse groups are often definable. For example, agricultural workers (the heterogeneous group discussed in the reasoning of  Polewsky ) may include all sorts of individuals (young, old, different religions, different races) who are still defined by a common occupation. Similarly, Lederer J appears to concede that a narrower conception of homelessness would also be definable. He qualifies that homelessness is indefinable “in the circumstances of this Application” (at para 129) because it is defined as having “inadequate housing” as opposed to having no housing at all. By implication, he suggests that if the homeless were simply defined as having no housing at all then such a group would be definable. Yet, such a group also encompasses a broad and diverse set of people. While the larger and more diverse a group becomes, the more difficult it may be to find a common thread tying them together, such a thread may nonetheless exist. As a consequence, I would argue that Lederer J improperly conflates indefinability with heterogeneity and ends up misapplying what is, in the jurisprudence, a relevant factor when assessing analogous grounds: status as a “discrete and insular minority” (see e.g. Andrews v Law Society of British Columbia, [1989] 1 SCR 143, 56 DLR (4th) 1 at 152 [Andrews] and Miron v Trudel, [1995] 2 SCR 418, 124 DLR (4th) 693 at para 148 (QL) [Miron])).

(c) Why Neither Definability nor Heterogeneity are Conclusive of the Analogous Grounds

It is unclear whether Lederer J rejects homelessness as an analogous ground because it is indefinable, heterogeneous, or both. Regardless, an analysis of only these two factors is incomplete because definability is irrelevant to the question of analogous grounds and heterogeneity is only one of many factors to be considered.

Definability cannot be a factor when assessing analogous grounds because it is not an attribute of the enumerated grounds. The basic analytical approach to analogous grounds is the drawing of analogies from the enumerated grounds for the purpose of identifying new sufficiently similar grounds worthy of protection under section 15 (see Corbiere v Canada (Minister of Indian and Northern Affairs), [1999] 2 SCR 203, 173 DLR (4th) 1 at para 13 [Corbiere]). Analogousness is an attribute which is intrinsically relative. To describe something as analogous begs the question: analogous to what? In the case of analogous grounds under the Charter, the legal question is whether or not the potential ground is analogous to the enumerated grounds. Thus, any consideration when identifying analogous grounds must be an attribute which attaches to the enumerated grounds.

Many enumerated grounds lack definability. For example, suppose a law discriminated against black people (that is, based on the enumerated ground of race). How would one qualify for membership within that group? A certain threshold of dark skin? A certain ethnic background? Other attributes classically (read stereotypically) associated with black people? Just like homelessness, the “parameters” that define what it means to be black (or any race for that matter) are not defined (unsurprisingly, given the constructed nature of race: see Ian F Haney López, “The Social Construction of Race” in Richard Delgado, ed, Critical Race Theory: The Cutting Edge (Philadelphia: Temple University Press, 1995) 191 available here). As another example, consider a government action that discriminated against Muslims (the enumerated ground of religion). Are all different denominations of Islam included? What if someone is non-practicing? What if different denominations have conflicting definitions of who fits under the umbrella of Muslim? Again, being Muslim may also lack clear parameters.

Admittedly, my characterization of definability is not identical to Lederer J’s. He considered homelessness indefinable because, in part, lacking “adequate housing” must be assessed on an “individual basis” (at para 129). In other words, depending on the size and make up of a particular family, more or less housing may be required for it to qualify as “adequate” housing. Regardless, his ultimate concern (unclear parameters, unclear membership) equally applies to multiple enumerated grounds. As a consequence, definability cannot be a requirement of the analogous grounds.

Heterogeneity is also not decisive of the analogous grounds, though it may be a relevant factor. Lederer J appears to rely on the notion of “discrete and insular minorities” for his consideration of heterogeneity (at para 136).

Whether or not a ground encompasses a “discrete and insular minority” (which is arguably the opposite of a disparate and heterogeneous group) has been considered by the courts (see e.g. Andrews, at 152 and Miron, at para 148 (QL)). However, this factor alone cannot be a gatekeeper to the analogous grounds. For example, in Andrews and Miron, whether or not the grounds in question encompassed a discrete and insular minority was considered along with other factors when determining analogousness. Furthermore, the limited weight of heterogeneity becomes apparent with a consideration of the currently protected grounds. For example, the first recognized analogous ground was citizenship (in Andrews). In that case, the Supreme Court held that it was contrary to the Charter to discriminate against “non-citizens” with respect to qualifying as legal professionals. Yet, non-citizens, just like the homeless, are a heterogeneous group. The same can be said for all of the enumerated grounds. Race (any type), age (at any stage of life), and sex (male, female, intersex, etc) all intersect with various other identities throughout Canadian society. All of these groups (with minor restrictions, like 5 year olds not being parents) would contain all of the “women, single mothers, persons with mental and physical disabilities, Aboriginal persons, seniors, youth, racialized persons,” etc that Lederer J referred to when rejecting homelessness as an analogous ground because of its diverse membership (at para 135).

While representing a “discrete and insular minority” is an aspect of analogous grounds, it is merely a factor to be considered. Additionally, while it is beyond the scope of this comment, the concept of “discrete and insular minorities” does not appear to have the same meaning that Lederer J assigns to it, namely, a small and homogenous group. In particular, when commenting on early jurisprudence that considered “discrete and insular minorities,” Dale Gibson interpreted this factor as relating to historical disadvantage which, if anything, furthers the argument for homelessness as an analogous ground (see Dale Gibson, “Analogous Grounds of Discrimination Under the Canadian Charter: Too Much Ado About Next to Nothing” (1991) 29 Alta L Rev 772 (QL)).

(2) A Multi-Variable Approach to Homelessness as an Analogous Ground

Generally, analogous grounds are described as personal characteristics that are either (1) immutable, like national origin, or (2) constructively immutable, like religion (Corbiere, at para 13). However, this tidy dichotomy overlooks the complex multi-variable analysis the Supreme Court often applies.

In “Assessing Analogous Grounds: The Doctrinal and Normative Superiority of a Multi-Variable Approach,” (2013) 10 JL & Equality 37 (HeinOnline) [“Assessing Analogous Grounds”], I argue that the existing jurisprudence surrounding section 15 goes beyond the concept of immutability and is open to many different variables such as those outlined in L’Heureux-Dubé J’s multi-variable approach from her concurring reasons in Corbiere. As a consequence, I will consider the ground of “homelessness” in light of immutability, constructive immutability, and several other indicia that have been applied by the Supreme Court in Charter jurisprudence when characterizing analogous grounds: difficulty and cost of change, vulnerability, historical disadvantage, and presence of the ground in human rights codes.

A multi-variable approach to analogous grounds favours the recognition of homelessness as an analogous ground.

Homelessness is not, strictly speaking, immutable. However, whether or not homelessness satisfies the test for “immutability” applied by the Supreme Court does not have a clear answer. As I argued in “Assessing Analogous Grounds”:

L’Heureux-Dubé J states in Egan that “the common characteristic of all of the enumerated grounds other than religion is that they involve so-called ‘immutable’ characteristics.” However, how would the Court characterize the actual immutability of generally stable, yet changeable, enumerated grounds such as sex and colour? How would the Court characterize the immutability of enumerated grounds that change over time but often due to factors beyond our control, such as age and temporary disability? (at 49).

Put differently, for the court to characterize the enumerated grounds (except for religion) as immutable means that the court has a flexible understanding of what it means for something to be “immutable.” Most notably:

[T]he Court has at times allowed for a very flexible and thus unpredictable understanding of immutability. For example, McLachlin J in Miron describes marital status as immutable “albeit in an attenuated form” because “it often lies beyond the individual’s effective control” (“Assessing Analogous Grounds,” at 49).

As a consequence, the analysis of the immutability of homelessness must look beyond the fact that it is subject to change (like sex and colour). That is not to say that a sex-change operation or skin bleaching is necessarily as common as escaping or entering homelessness. That being said, if “often [lying] beyond the individual’s effective control” can qualify as an “attenuated” form of immutability, then homelessness could very well meet that standard.

Whether homelessness is constructively immutable is also unclear. The leading decision on constructive immutability is Corbiere, which held that a personal characteristic is constructively immutable if (1) it is changeable only at unacceptable cost to personal identity, or (2) the government has no legitimate interest in expecting an individual to change the personal characteristic (see “Assessing Analogous Grounds,” at 40-41). Such an approach, which implicitly assumes that personal characteristics are desirable, is confusing when they are not desirable. As I discussed in the context of the potential ground of “drug addiction”:

Many potential grounds of discrimination struggle to satisfy this test. For example, escaping drug addiction is possible (though “very difficult”) and becoming someone who is no longer an addict does not result in unacceptable cost to personal identity. If anything, being a drug addict is the cost and escaping addiction the benefit (“Assessing Analogous Grounds,” at 52-53).

Homelessness similarly struggles to fit within this framework of constructive immutability.