Dunmore v. Ontario (Attorney General), [2001] 3 S.C.R. 1016, 2001 SCC 94

Tom Dunmore, Salame Abdulhamid,

Walter Lumsden and Michael Doyle, on their

own behalf and on behalf of the United Food and

Commercial Workers International Union Appellants

v.

Attorney General for Ontario

and Fleming Chicks Respondents

and

Attorney General of Quebec,

Attorney General for Alberta,

Canadian Labour Congress and

Labour Issues Coordinating Committee (“LICC”) Interveners

Indexed as: Dunmore v. Ontario (Attorney General)

Neutral citation: 2001 SCC 94.

File No.: 27216.

2001: February 19; 2001: December 20.

Present: McLachlin C.J. and L’Heureux‑Dubé, Gonthier, Iacobucci, Major, Bastarache, Binnie, Arbour and LeBel JJ.

on appeal from the court of appeal for ontario

Constitutional law -- Charter of Rights -- Freedom of association -- Exclusion of agricultural workers from statutory labour relations regime -- Whether exclusion infringes freedom of association – If so, whether infringement justifiable -- Canadian Charter of Rights and Freedoms, ss. 1, 2(d) -- Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, s. 80 -- Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, s. 3(b).

In 1994, the Ontario legislature enacted the Agricultural Labour Relations Act, 1994 (“ALRA”), which extended trade union and collective bargaining rights to agricultural workers. Prior to the adoption of this legislation, agricultural workers had always been excluded from Ontario’s labour relations regime. A year later, by virtue of s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995 (“LRESLAA”), the legislature repealed the ALRA in its entirety, in effect subjecting agricultural workers to s. 3(b) of the Labour Relations Act, 1995 (“LRA”), which excluded them from the labour relations regime set out in the LRA. Section 80 also terminated any certification rights of trade unions, and any collective agreements certified, under the ALRA. The appellants brought an application challenging the repeal of the ALRA and their exclusion from the LRA, on the basis that it infringed their rights under ss. 2(d) and 15(1) of the Canadian Charter of Rights and Freedoms. Both the Ontario Court (General Division) and the Ontario Court of Appeal upheld the challenged legislation.

Held (Major J. dissenting): The appeal should be allowed. The impugned legislation is unconstitutional.

Per McLachlin C.J. and Gonthier, Iacobucci, Bastarache, Binnie, Arbour and LeBel JJ.: The purpose of s. 2(d) of the Charter is to allow the achievement of individual potential through interpersonal relationships and collective action. This purpose commands a single inquiry: has the state precluded activity because of its associational nature, thereby discouraging the collective pursuit of common goals? While the traditional four-part formulation of the content of freedom of association sheds light on this concept, it does not capture the full range of activities protected by s. 2(d). In some cases s. 2(d) should be extended to protect activities that are inherently collective in nature, in that they cannot be performed by individuals acting alone. Trade unions develop needs and priorities that are distinct from those of their members individually and cannot function if the law protects exclusively the lawful activities of individuals. The law must thus recognize that certain union activities may be central to freedom of association even though they are inconceivable on the individual level.

Ordinarily, the Charter does not oblige the state to take affirmative action to safeguard or facilitate the exercise of fundamental freedoms. There is no constitutional right to protective legislation per se. However, history has shown and Canada’s legislatures have recognized that a posture of government restraint in the area of labour relations will expose most workers not only to a range of unfair labour practices, but potentially to legal liability under common law inhibitions on combinations and restraints of trade. In order to make the freedom to organize meaningful, in this very particular context, s. 2(d) of the Charter may impose a positive obligation on the state to extend protective legislation to unprotected groups. The distinction between positive and negative state obligations ought to be nuanced in the context of labour relations, in the sense that excluding agricultural workers from a protective regime contributes substantially to the violation of protected freedoms.

Several considerations circumscribe the possibility of challenging underinclusion under s. 2 of the Charter: (1) claims of underinclusion should be grounded in fundamental Charter freedoms rather than in access to a particular statutory regime; (2) the evidentiary burden in cases where there is a challenge to underinclusive legislation is to demonstrate that exclusion from a statutory regime permits a substantial interference with the exercise of protected s. 2(d) activity; and (3), in order to link the alleged Charter violation to state action, the context must be such that the state can be truly held accountable for any inability to exercise a fundamental freedom. The contribution of private actors to a violation of fundamental freedoms does not immunize the state from Charter review.

In order to establish a violation of s. 2(d) of the Charter, the appellants must demonstrate that their claim relates to activities that fall within the range of activities protected by s. 2(d) of the Charter, and that the impugned legislation has, either in purpose or effect, interfered with these activities. In this case, insofar as the appellants seek to establish and maintain an association of employees, their claim falls squarely within the protected ambit of s. 2(d). Moreover, the effective exercise of the freedoms in s. 2(d) require not only the exercise in association of the constitutional rights and freedoms and lawful rights of individuals, but the exercise of certain collective activities, such as making majority representations to one’s employer. Conflicting claims concerning the meaning of troubling comments in the legislature make it impossible to conclude that the exclusion of agricultural workers from the LRA was intended to infringe their freedom to organize, but the effect of the exclusion in s. 3(b) of the LRA is to infringe their right to freedom of association.

The LRA is clearly designed to safeguard the exercise of the freedom to associate rather than to provide a limited statutory entitlement to certain classes of citizens. Through the right to organize inscribed in s. 5 of the LRA and the protection offered against unfair labour practices, the legislation recognizes that without a statutory vehicle employee associations are, in many cases, impossible. Here, the appellants do not claim a constitutional right to general inclusion in the LRA, but simply a constitutional freedom to organize a trade association. This freedom to organize exists independently of any statutory enactment, although its effective exercise may require legislative protection in some cases. The appellants have met the evidentiary burden of showing that they are substantially incapable of exercising their fundamental freedom to organize without the LRA’s protective regime. While the mere fact of exclusion from protective legislation is not conclusive evidence of a Charter violation, the evidence indicates that, but for the brief period covered by the ALRA, there has never been an agricultural workers’ union in Ontario and agricultural workers have suffered repeated attacks on their efforts to unionize. The inability of agricultural workers to organize can be linked to state action. The exclusion of agricultural workers from the LRA functions not simply to permit private interferences with their fundamental freedoms, but to substantially reinforce such interferences. The inherent difficulties of organizing farm workers, combined with the threat of economic reprisal from employers, form only part of the reason why association is all but impossible in the agricultural sector in Ontario. Equally important is the message sent by the exclusion of agricultural workers from the LRA, which delegitimizes their associational activity and thereby contributes to its ultimate failure. The most palpable effect of the LRESLAA and the LRA is, therefore, to place a chilling effect on non-statutory union activity.

With respect to the s. 1 analysis, the evidence establishes that many farms in Ontario are family-owned and operated, and that the protection of the family farm is a pressing enough objective to warrant the infringement of s. 2(d) of the Charter. The economic objective of ensuring farm productivity is also important. Agriculture occupies a volatile and highly competitive part of the private sector economy, experiences disproportionately thin profit margins and, due to its seasonal character, is particularly vulnerable to strikes and lockouts.

There is also a rational connection between the exclusion of agricultural workers from Ontario’s labour relations regime and the objective of protecting the family farm. Unionization leads to formalized labour-management relationships and gives rise to a relatively formal process of negotiation and dispute resolution. It is reasonable to speculate that unionization will threaten the flexibility and cooperation that is characteristic of the family farm. Yet this concern is only as great as the extent of the family farm structure in Ontario and does not necessarily apply to the right to form an agricultural association. The notion that employees should sacrifice their freedom to associate in order to maintain a flexible employment relationship should be carefully circumscribed, as it could, if left unchecked, justify restrictions on unionization in many sectors of the economy.

The wholesale exclusion of agricultural workers from Ontario’s labour relations regime does not minimally impair their right to freedom of association. The categorical exclusion of agricultural workers is unjustified where no satisfactory effort has been made to protect their basic right to form associations. The exclusion is overly broad as it denies the right of association to every sector of agriculture without distinction. The reliance on the family farm justification ignores an increasing trend in Canada towards corporate farming and complex agribusiness and does not justify the unqualified and total exclusion of all agricultural workers from Ontario’s labour relations regime. More importantly, no justification is offered for excluding agricultural workers from all aspects of unionization, in particular those protections that are necessary for the effective formation and maintenance of employee associations. Nothing in the record suggests that protecting agricultural workers from the legal and economic consequences of forming an association would pose a threat to the family farm structure. Consequently, the total exclusion of agricultural workers from Ontario’s labour relations regime is not justifiable under s. 1 of the Charter.

The appropriate remedy in this case is to declare the LRESLAA unconstitutional to the extent that it gives effect to the exclusion clause found in s. 3(b) of the LRA, and to declare s. 3(b) of the LRA unconstitutional. The declarations should be suspended for 18 months, thereby allowing amending legislation to be passed if the legislature sees fit to do so. Section 2(d) of the Charter only requires the legislature to provide a statutory framework that is consistent with the principles established in this case. At a minimum, these principles require that the statutory freedom to organize in s. 5 of the LRA be extended to agricultural workers, along with protections judged essential to its meaningful exercise, such as freedom to assemble, freedom from interference, coercion and discrimination and freedom to make representations and to participate in the lawful activities of the association. The appropriate remedy does not require or forbid the inclusion of agricultural workers in a full collective bargaining regime, whether it be the LRA or a special regime applicable only to agricultural workers.

It is unnecessary to consider the status of occupational groups under s. 15(1) of the Charter.

Per L’Heureux-Dubé J.: The purpose of s. 80 of the LRESLAA and s. 3(b) of the LRA is to prevent agricultural workers from unionizing, and this purpose infringes s. 2(d) of the Charter. In the record, there is clear evidence of intent on the part of the government of Ontario to breach the s. 2(d) rights of agricultural workers, including repeated instances where government officials indicated that the impugned legislation’s intent was to hinder union-related activities in the agricultural sector. On a balance of probabilities, the evidence demonstrates that the legislature’s purpose in enacting the exclusion was to ensure that persons employed in agriculture remained vulnerable to management interference with their associational activities, in order to prevent the undesirable consequences which it had feared would result from agricultural workers’ labour associations. Furthermore, the evidence does not reveal any positive effects upon the associational freedom of agricultural workers stemming from their exclusion from the LRA. The reality of the labour market, which has led to the development of protective labour legislation, indicates that when the protection is removed without any restriction or qualification, associational rights are often infringed, or have the potential to be infringed, to an extent not confined to unionization activities. Consequently, it was in the reasonable contemplation of the government at the time of the enactment of the impugned legislation that the effect of the exclusion clause would be to affect associational freedoms beyond the realm of unionization, thus breaching s. 2(d) Charter rights.

In the present case, there is a positive obligation on the government to provide legislative protection against unfair labour practices. A positive duty to assist excluded groups generally arises when the claimants are in practice unable to exercise a Charter right. In the case of agricultural workers in Ontario, the freedom to associate becomes meaningless in the absence of a duty of the State to take positive steps to ensure that this right is not a hollow one. The government has breached the s. 2(d) rights of agricultural workers because it has enacted a new labour statute which leaves them perilously vulnerable to unfair labour practices. The absolute removal of LRA protection from agricultural workers has created a situation where employees have reason to fear retaliation against associational activity by employers. In light of the reality of the labour market, the failure of the Ontario legislature to spell out a regime defining which associational activities are to be protected from management retaliation has a chilling effect on freedom of association for agricultural workers. The chilling effect of the impugned provision has forced agricultural workers to abandon associational efforts and restrain themselves from further associational initiatives. The freedom of association of agricultural workers under the LRA can be characterized as a hollow right because it amounts to no more than the freedom to suffer serious adverse legal and economic consequences. In a constitutional democracy, not only must fundamental freedoms be protected from State action, they must also be given “breathing space”.

Since the impugned legislation infringes s. 2(d), it is necessary to make but a single observation with respect to whether the exclusion of agricultural workers from the LRA constitutes discrimination under s. 15(1) of the Charter. The occupational status of agricultural workers constitutes an “analogous ground” for the purposes of an analysis under s. 15(1). There is no reason why an occupational status cannot, in the right circumstances, identify a protected group. Employment is a fundamental aspect of an individual’s life and an essential component of identity, personal dignity, self-worth and emotional well-being. Agricultural workers generally suffer from disadvantage and the effect of the distinction made by their exclusion from the LRA is to devalue and marginalize them within Canadian society. Agricultural workers, in light of their relative status, low levels of skill and education, and limited employment and mobility, can change their occupational status only at great cost, if at all.

The impugned legislation is not justifiable under s. 1 of the Charter. While labour statutes, such as the LRA, fulfill important objectives in our society, s. 3(b) does not pursue a pressing and substantial concern justifying the breach of the appellants’ Charter rights. It cannot be argued that Ontario agriculture has unique characteristics which are incompatible with legislated collective bargaining. It is also difficult to accept that none of the LRA’s purposes, enumerated at s. 2 of the LRA, which speak to the basic characteristics required for the operation of a modern business, are inapplicable in the agricultural sector. At the very least, the expressions of intent found in s. 2 of the LRA would apply to factory-like agricultural enterprises. Without enunciating a constitutionally valid reason, one cannot countenance a breach of a Charter guaranteed fundamental freedom on grounds which appear to be based on a policy geared to enhance the economic well-being of private enterprises. The government is entitled to provide financial and other support to agricultural operations, including family farms. However, it is not open to the government to do so at the expense of the Charter rights of those who are employed in such activities, if such a policy choice cannot be demonstrably justified.

Even if the impugned legislation pursued a valid objective, the absoluteness of the exclusion clause, barring all persons employed in agriculture from all components of the LRA, speaks to the lack of proportionality between the perceived ills to be avoided and their remedy. First, a rational connection between the objective of securing the well-being of the agricultural sector in Ontario and the exclusion of persons employed in agriculture from all associational protections contained in the LRA has not been established. If the good labour management principles outlined in s. 2 of the LRA have a basis in fact, then barring all persons employed in agriculture from all the benefits under the LRA may have the opposite effect. Second, the complete exclusion of agricultural workers from the LRA does not minimally impair their Charter rights. Such a blunt measure can hardly be characterized as achieving a delicate balance among the interests of labour and those of management and the public. It weakens the case for deference to the legislature. This is further aggravated because those affected by the exclusion are not only vulnerable as employees but are also vulnerable as members of society with low income, little education and scant security or social recognition. The current law is not carefully tailored to balance the Charter freedoms of persons employed in agriculture in Ontario and the societal interest in harmonious relations in the labour market. While the important role that family farms play in Ontario agriculture must be recognized, such a role is not unique to Ontario. Further, both families and farms have evolved. There is no obvious connection between the exclusion of agricultural workers from the LRA and farmers or family farms. A city-based corporation could be operating an agricultural entity and benefit from the restrictions on the freedoms of association of its agricultural workers. Labour statutes in other provinces contain agricultural exemptions that are narrower than the one contained in the LRA. The objective of securing the well-being of the agricultural sector in Ontario can be achieved through a legislative mechanism that is less restrictive of free association than the existing complete exclusion of agricultural workers from the LRA.

Per Major J. (dissenting): The appellants failed to demonstrate that the impugned legislation has, either in purpose or effect, infringed activities protected by s. 2(d) of the Charter. In particular, s. 2(d) does not impose a positive obligation of protection or inclusion on the state in this case. Prior to the enactment of the LRA, agricultural workers had historically faced significant difficulties organizing and the appellants did not establish that the state is causally responsible for the inability of agricultural workers to exercise a fundamental freedom.

Agricultural workers are not an analogous group for the purposes of s. 15(1) of the Charter and, as a result, the exclusion of agricultural workers from the LRA does not violate their equality rights.

Cases Cited

By Bastarache J.

Referred to: Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989; RWDSU v. Dolphin Delivery Ltd., 1986 CanLII 5 (SCC), [1986] 2 S.C.R. 573; Ferrell v. Ontario (Attorney General) (1997), 1997 CanLII 12351 (ON SC), 149 D.L.R. (4th) 335; R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927; Reference re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313; PSAC v. Canada, 1987 CanLII 89 (SCC), [1987] 1 S.C.R. 424; RWDSU v. Saskatchewan, 1987 CanLII 90 (SCC), [1987] 1 S.C.R. 460; Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), 1990 CanLII 72 (SCC), [1990] 2 S.C.R. 367; Canadian Egg Marketing Agency v. Richardson, 1997 CanLII 17020 (SCC), [1998] 3 S.C.R. 157; Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211; R. v. Advance Cutting & Coring Ltd., [2001] 3 S.C.R. 209, 2001 SCC 70; R. v. Skinner, 1990 CanLII 107 (SCC), [1990] 1 S.C.R. 1235; Syndicat catholique des employés de magasins de Québec Inc. v. Compagnie Paquet Ltée, 1959 CanLII 51 (SCC), [1959] S.C.R. 206; McGavin Toastmaster Ltd. v. Ainscough, 1975 CanLII 9 (SCC), [1976] 1 S.C.R. 718; R. v. Beaulac, 1999 CanLII 684 (SCC), [1999] 1 S.C.R. 768; Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995; Native Women’s Assn. of Canada v. Canada, 1994 CanLII 27 (SCC), [1994] 3 S.C.R. 627; R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713; Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493; Law v. Canada (Minister of Employment and Immigration), 1999 CanLII 675 (SCC), [1999] 1 S.C.R. 497; McKinney v. University of Guelph, 1990 CanLII 60 (SCC), [1990] 3 S.C.R. 229; Machtinger v. HOJ Industries Ltd., 1992 CanLII 102 (SCC), [1992] 1 S.C.R. 986; Wallace v. United Grain Growers Ltd., 1997 CanLII 332 (SCC), [1997] 3 S.C.R. 701; R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103; Thomson Newspapers Co. v. Canada (Attorney General), 1998 CanLII 829 (SCC), [1998] 1 S.C.R. 877; R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2; Wellington Mushroom Farm, [1980] O.L.R.B. Rep. May 813; Calvert-Dale Estates Ltd., [1971] O.L.R.B. Rep. Feb. 58; Spruceleigh Farms, [1972] O.L.R.B. Rep. Oct. 860; Cuddy Chicks Ltd., [1988] O.L.R.B. Rep. May 468, application for judicial review dismissed (1988), 1988 CanLII 4839 (ON SC), 66 O.R. (2d) 284, aff’d (1989), 1989 CanLII 4139 (ON LRB), 70 O.R. (2d) 179, aff’d 1991 CanLII 57 (SCC), [1991] 2 S.C.R. 5; Osborne v. Canada (Treasury Board), 1991 CanLII 60 (SCC), [1991] 2 S.C.R. 69; Rocket v. Royal College of Dental Surgeons of Ontario, 1990 CanLII 121 (SCC), [1990] 2 S.C.R. 232; South Peace Farms and Oil, Chemical and Atomic Workers International Union, Local No. 9-686, [1977] 1 Can. L.R.B.R. 441; Rodriguez v. British Columbia (Attorney General), 1993 CanLII 75 (SCC), [1993] 3 S.C.R. 519.

By L’Heureux-Dubé J.

Referred to: Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995; Reference re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313; Native Women’s Assn. of Canada v. Canada, 1994 CanLII 27 (SCC), [1994] 3 S.C.R. 627; Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989; Cuddy Chicks Ltd. v. Ontario (Labour Relations Board), 1991 CanLII 57 (SCC), [1991] 2 S.C.R. 5; Cuddy Chicks Ltd., [1988] O.L.R.B. Rep. May 468; Cuddy Chicks Ltd., [1992] O.L.R.D. No. 1170 (QL); New Brunswick Broadcasting Co. v. Nova Scotia (Speaker of the House of Assembly), 1993 CanLII 153 (SCC), [1993] 1 S.C.R. 319; Hunter v. Southam Inc., 1984 CanLII 33 (SCC), [1984] 2 S.C.R. 145; R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103; R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295, aff’g (1983), 1983 ABCA 268, 49 A.R. 194; Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493; Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211; R. v. Advance Cutting & Coring Ltd., [2001] 3 S.C.R. 209, 2001 SCC 70; Egan v. Canada, 1995 CanLII 98 (SCC), [1995] 2 S.C.R. 513; R. v. Beaulac, 1999 CanLII 684 (SCC), [1999] 1 S.C.R. 768; Committee for the Commonwealth of Canada v. Canada, 1991 CanLII 119 (SCC), [1991] 1 S.C.R. 139; Osborne v. Canada (Treasury Board), 1991 CanLII 60 (SCC), [1991] 2 S.C.R. 69; Law v. Canada (Minister of Employment and Immigration), 1999 CanLII 675 (SCC), [1999] 1 S.C.R. 497; Corbiere v. Canada (Minister of Indian and Northern Affairs), 1999 CanLII 687 (SCC), [1999] 2 S.C.R. 203; McKinley v. BC Tel, [2001] 2 S.C.R. 161, 2001 SCC 38; R. v. Généreux, 1992 CanLII 117 (SCC), [1992] 1 S.C.R. 259; Andrews v. Law Society of British Columbia, 1989 CanLII 2 (SCC), [1989] 1 S.C.R. 143; R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713; RJR-MacDonald Inc. v. Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 S.C.R. 199; Ross v. New Brunswick School District No. 15, 1996 CanLII 237 (SCC), [1996] 1 S.C.R. 825; South Peace Farms and Oil, Chemical and Atomic Workers International Union, Local No. 9-686, [1977] 1 Can. L.R.B.R. 441; Wellington Mushroom Farm, [1980] O.L.R.B. Rep. May 813; Rathwell v. Rathwell, 1978 CanLII 3 (SCC), [1978] 2 S.C.R. 436.

By Major J. (dissenting)

Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989; Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995.

Statutes and Regulations Cited

Agricultural Labour Relations Act, 1994, S.O. 1994, c. 6 [rep. 1995, c. 1, s. 80], s. 10.

Canadian Charter of Rights and Freedoms, ss. 1, 2(a), (b), (d), 15(1), 32.

Collective Bargaining Act, 1943, S.O. 1943, c. 4, s. 24 [rep. & sub. by Labour Relations Board Act, 1944, S.O. 1944, c. 29].

Convention (No. 11) concerning the Rights of Association and Combination of Agricultural Workers, 38 U.N.T.S. 153, Art. 1.

Convention (No. 87) concerning Freedom of Association and Protection of the Right to Organize, 67 U.N.T.S. 17, Arts. 2, 9, 10.

Convention (No. 141) concerning Organisations of Rural Workers and Their Role in Economic and Social Development, I.L.O. Official Bulletin, vol. LVIII, 1975, Series A, No. 1, p. 28, Art. 2.

Corporations Tax Act, R.S.O. 1990, c. C.40, s. 1(2) “family farm corporation”.

Employment Standards Act, R.S.O. 1990, c. E.14.

Employment Standards Act Regulations, R.R.O. 1990, Reg. 325, s. 3(1)(i).

Fish and Wildlife Conservation Act, 1997, S.O. 1997, c. 41, s. 1(1) “farmer”.

Industrial Relations Act, R.S.N.B. 1973, c. I-4, s. 1(5)(a).

Junior Farmer Establishment Act, R.S.O. 1990, c. J.2, s. 1 “family farm”.

Labour Code, R.S.Q., c. C-27, s. 21.

Labour Code of British Columbia Act, S.B.C. 1973, c. 122.

Labour Relations Act, R.S.O. 1980, c. 228, s. 2(b).

Labour Relations Act, 1950, S.O. 1950, c. 34, s. 2(b) [am. 1960, c. 54, s. 2(bb)].

Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, ss. 1(1) “agriculture”, 2, 3, 3(b), (e) [am. 1997, c. 4, s. 83], (f) [idem, c. 31, s. 151], 5, 13, 70, 72, 76, 86, 87, 88.

Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, ss. 80, 81(1).

Labour Relations Board Act, 1944, S.O. 1944, c. 29, s. 10(a) [rep. & sub. Labour Relations Act, 1948, S.O. 1948, c. 51].

National Labor Relations Act, July 5, 1935, c. 372, 49 Stat. 449 (29 U.S.C. §§ 151 to 169).

Occupational Health and Safety Act, R.S.O. 1990, c. O.1, s. 3(2).

Tenant Protection Act, 1997, S.O. 1997, c. 24, s. 3(b).

Authors Cited

Adams, George W. Canadian Labour Law, 2nd ed. Aurora, Ont.: Canada Law Book (loose-leaf updated June 2001, release 15).

Arthurs, H. W., et al. Labour Law and Industrial Relations in Canada, 4th ed. Markham, Ont.: Butterworths, 1993.

Beatty, David M. Putting the Charter to Work: Designing a Constitutional Labour Code. Kingston, Ont.: McGill-Queen’s University Press, 1987.

Beatty, David, and Steven Kennett. “Striking Back: Fighting Words, Social Protest and Political Participation in Free and Democratic Societies” (1988), 67 Can. Bar Rev. 573.

Canada. Parliament. Special Joint Committee on the Constitution of Canada. Minutes of Proceedings and Evidence of the Special Joint Committee of the Senate and House of Commons on the Constitution of Canada, Issue No. 43, January 22, 1981, pp. 69-70.

Canada. Statistics Canada. Industrial Organization and Finance Division. Labour Unions Section. Annual Report of the Minister of Industry, Science and Technology under the Corporations and Labour Unions Returns Act, Part II, Labour Unions. Ottawa: Statistics Canada, 1992.

Canada. Task Force on Labour Relations. Canadian Industrial Relations: The Report of the Task Force on Labour Relations. Ottawa: Privy Council Office, 1968.

Gibson, Dale. The Law of the Charter: Equality Rights. Toronto: Carswell, 1990.

Harmer, Lily. “The Right to Strike: Charter Implications and Interpretations” (1988), 47 U.T. Fac. L. Rev. 420.

Hodges-Aeberhard, Jane. “The right to organise in Article 2 of Convention No. 87: What is meant by workers ‘without distinction whatsoever’?” (1989), 128 Int’l Lab. Rev. 177.

Hogg, Peter W. Constitutional Law of Canada, vol. 2, loose-leaf ed. Scarborough, Ont.: Carswell, 1992 (updated 2000, release 1).

Hunter, Ian. “Individual and Collective Rights in Canadian Labour Law” (1993), 22 Man. L.J. 145.

International Labour Office. Committee on Freedom of Association. Report 308. Case No. 1900 (Canada/Ontario): Report in which the Committee requests to be kept informed of developments, I.L.O. Official Bulletin, vol. LXXX, 1997, Series B, No. 3.

International Labour Office. Freedom of Association Committee. Freedom of Association: Digest of decisions and principles of the Freedom of Association Committee of the Governing Body of the ILO, 4th ed. Geneva: International Labour Office, 1996.

International Labour Office. Voices for Freedom of Association, Labour Education 1998/3, No. 112. Geneva: International Labour Office, 1998.

Labour Law Casebook Group. Labour and Employment Law: Cases, Materials and Commentary, 6th ed. Kingston, Ont.: Industrial Relations Centre, Queen’s University, 1998.

Morris, Charles J., ed. The Developing Labor Law. Washington: Bureau of National Affairs, 1971.

Neilson, Kathryn, and Innis Christie. “The Agricultural Labourer in Canada: A Legal Point of View” (1975), 2 Dal. L.J. 330.

Ontario Government Media Kit for Bill 7 (An Act to Restore Balance and Stability to Labour Relations and to Promote Economic Prosperity). Questions and Answers: Repeal of the Agricultural Labour Relations Act (Bill 91). Toronto: Queen’s Park, 1995.

Ontario. Legislative Assembly. Official Report of Debates, 1st Sess., 36th Parl., October 4, 1995, pp. 99-100.

Ontario. Task Force on Agricultural Labour Relations. Report to the Minister of Labour, June 1992.

Ontario. Task Force on Agricultural Labour Relations. Second Report to the Minister of Labour, November 1992.

Pothier, Dianne. “Connecting Grounds of Discrimination to Real People’s Real Experiences” (2001), 13 C.J.W.L. 37.

Sugeno, Kazuo. “Unions as social institutions in democratic market economies” (1994), 133 Int’l Lab. Rev. 511.

Swepston, Lee. “Human rights law and freedom of association: Development through ILO supervision” (1998), 137 Int’l Lab. Rev. 169.

Weiler, Paul C. Reconcilable Differences: New Directions in Canadian Labour Law. Toronto: Carswell, 1980.

White, James. A Profile of Ontario Farm Labour. Brampton, Ont.: James White & Associates, 1997.

APPEAL from a judgment of the Ontario Court of Appeal (1999), 1999 CanLII 18653 (ON CA), 182 D.L.R. (4th) 471, 49 C.C.E.L. (2d) 29, [1999] O.J. No. 1104 (QL), upholding the decision of the Ontario Court (General Division) (1997), 1997 CanLII 16229 (ON SC), 155 D.L.R. (4th) 193, 37 O.R. (3d) 287, 49 C.C.E.L. (2d) 5, [1997] O.J. No. 4947 (QL). Appeal allowed, Major J. dissenting.

Chris G. Paliare and Martin J. Doane, for the appellants.

Richard J. K. Stewart, for the respondent the Attorney General for Ontario.

Alan L. W. D’Silva, Darrell L. Kloeze and Vincent C. Kazmierski, for the respondent Fleming Chicks.

Written submissions only by Renée Madore and Monique Rousseau, for the intervener the Attorney General of Quebec.

Rod Wiltshire, for the intervener the Attorney General for Alberta.

Steven Barrett, for the intervener the Canadian Labour Congress.

Written submissions only by John C. Murray and Jonathan L. Dye, for the intervener the Labour Issues Coordinating Committee.

The judgment of McLachlin C.J. and Gonthier, Iacobucci, Bastarache, Binnie, Arbour and LeBel JJ. was delivered by

Bastarache J. --

I. Introduction

1 This appeal concerns the exclusion of agricultural workers from Ontario’s statutory labour relations regime. The appellants, individual farm workers and union organizers, challenge the exclusion as a violation of their freedom of association and equality rights under the Canadian Charter of Rights and Freedoms. In particular, they argue that the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1 (“LRESLAA”), combined with s. 3(b) of the Ontario Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A (“LRA”), prevents them from establishing, joining and participating in the lawful activities of a trade union. In addition, they claim that the LRESLAA and the LRA violate their equality rights under s. 15(1) of the Charter by denying them a statutory protection enjoyed by most occupational groups in Ontario.

2 This is the first time this Court has been asked to review the total exclusion of an occupational group from a statutory labour relations regime, where that group is not employed by the government and has demonstrated no independent ability to organize. For this reason, this appeal raises novel issues of state responsibility under s. 2(d) of the Charter, notwithstanding its apparent similarity to recent labour relations jurisprudence. After considering these issues, I conclude that the total exclusion of agricultural workers from the LRA violates s. 2(d) of the Charter and cannot be justified under s. 1. Accordingly, I conclude that, at a minimum, whatever protections are necessary to establish and maintain employee associations should be extended to persons employed in agriculture in Ontario. I am also of the view that it is not necessary to consider the status of agricultural workers under s. 15(1) of the Charter; assuming without deciding the existence of a s. 15(1) violation, such a violation would not alter the remedy I propose.

II. Factual Background

3 Although agricultural workers have been excluded from Ontario’s labour relations regime since 1943, the impetus for this appeal was the passage of the LRESLAA. The LRESLAA was enacted pursuant to an initiative of Ontario’s Progressive Conservative government in 1995; it repealed the only statute ever to extend trade union and collective bargaining rights to Ontario’s agricultural workers. That short-lived statute, the Agricultural Labour Relations Act, 1994, S.O. 1994, c. 6 (“ALRA”), was enacted pursuant to an initiative of the New Democratic Party government in 1994 following the recommendations in the Report of the Task Force on Agricultural Labour Relations: Report to the Minister of Labour (June 1992). The ALRA lasted from June 23, 1994 to November 10, 1995, during which time the United Food and Commercial Workers Union (“UFCW”) was certified as the bargaining agent for approximately 200 workers at the Highline Produce Limited mushroom factory in Leamington, Ontario. The UFCW also filed two other certification applications during the period of the ALRA, one for the workers at the Kingsville Mushroom Farm Inc., and the other for the workers at the respondent Fleming Chicks. These certification activities came to an end when, with the passage of the LRESLAA in 1995, the ALRA was repealed in its entirety. In addition to terminating any agreements certified under the ALRA, the LRESLAA terminated any certification rights of trade unions and prohibited employers from punishing workers for any union activity conducted under the ALRA. The appellants brought an application within one week of the repeal of the ALRA, arguing that it infringed their rights under ss. 2(d) and 15(1) of the Charter.

4 As indicated by the legislative record, the LRESLAA represents a small piece of the factual context surrounding this litigation. For over 50 years prior to the ALRA, and ever since its repeal, the Ontario government has excluded agricultural workers from its statutory labour relations regime. The first statute to effect this exclusion was the Collective Bargaining Act, 1943, S.O. 1943, c. 4, which was modelled on the American National Labor Relations Act (“Wagner Act”), July 5, 1935, c. 372, 49 Stat. 449 (29 U.S.C. §§ 151 to 169). Section 24 of the Collective Bargaining Act, 1943 contained a list of excluded classes, including “domestic servants”, “members of any police force”, certain other public employees, and “the industry of farming”. The most recent embodiment of Ontario’s labour relations policy, the LRA, excludes agricultural workers in the following terms:

3. This Act does not apply,

(b) to a person employed in agriculture, hunting or trapping;

The net effect of the LRESLAA was to re-subject agricultural workers to this exclusion clause. Thus, in addition to challenging the constitutionality of the LRESLAA, the appellants challenge the constitutionality of s. 3(b) of the LRA.

5 The various enactments dealing with agricultural workers’ right to unionize, the Wagner Act, the LRA, the ALRA and the LRESLAA, reflect highly divergent approaches to economic and labour policy. As noted by Sharpe J. (as he then was) in the Ontario Court (General Division), the current government in Ontario has “a very different perspective from that of its predecessor on appropriate economic and labour policy” and, indeed, rejects any attempt to include agricultural workers in its labour relations regime ((1997), 1997 CanLII 12345 (ON SC), 155 D.L.R. (4th) 193, at p. 199). Moreover, the affidavit evidence in this case “presents in stark contrast two conflicting views of an appropriate labour relations regime for agricultural workers in Ontario”, one denying the existence of any “industrial relations rationale” for the current exclusion, and the other maintaining that the collective bargaining model of the ALRA or the LRA would unduly threaten the province’s farm economy (pp. 201-2). This latter view is evidently shared by the Legislature of Alberta, which is the only other Canadian province to exclude agricultural workers from its labour relations regime. What is central to this appeal, however, is the constitutional effect of excluding agricultural workers from the LRA from the perspective of their freedom to associate. Given my conclusion that this exclusion violates s. 2(d) of the Charter, the above evidence will provide an important foundation for the s. 1 analysis.

III. Relevant Statutory Provisions

6 Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1

1.(1) The Labour Relations Act, 1995, as set out in Schedule A, is hereby enacted.

80.(1) The Agricultural Labour Relations Act, 1994 is repealed.

(2) On the day on which this section comes into force, a collective agreement ceases to apply to a person to whom that Act applied.

(3) On the day on which this section comes into force, a trade union certified under that Act or voluntarily recognized as the bargaining agent for employees to whom that Act applies ceases to be their bargaining agent.

(4) On the day on which this section comes into force, any proceeding commenced under that Act is terminated.

Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A

3. This Act does not apply,

(b) to a person employed in agriculture, hunting or trapping;

Canadian Charter of Rights and Freedoms

2. Everyone has the following fundamental freedoms:

(d) freedom of association.

IV. Judicial History

A. Ontario Court (General Division) (1997), [1997 CanLII 12345 (ON SC)](https://www.canlii.org/en/on/onsc/doc/1997/1997canlii12345/1997canlii12345.html), 155 D.L.R. (4th) 193

7 The issues before the Ontario Court (General Division) in this case were essentially the same as those before this Court, namely, whether the exclusion of agricultural workers from Ontario’s statutory labour relations scheme infringes s. 2(d) and/or s. 15(1) of the Charter and, if so, whether the infringements are justifiable under s. 1. It might be noted that Sharpe J. released his decision prior to this Court’s decisions in Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989, and other cases that provide important authority for this appeal.

8 Sharpe J. began with the appellants’ s. 2(d) claim, observing that the right of workers to form a trade union is protected by s. 2(d), while the right to collective bargaining is not. The balance of Sharpe J.’s s. 2(d) analysis was thus devoted to whether the impugned provisions infringed, either in purpose or effect, the former right. With respect to purpose, Sharpe J. held that while the purpose of the legislation was undoubtedly to deny agricultural workers the right to bargain collectively, “it is difficult . . . to discern a governmental purpose to deny agricultural workers the right to form an association” (pp. 205-6). He then considered the effect of the legislation on s. 2(d) rights, holding that to the extent agricultural workers are deprived of the ability to form trade unions, such deprivation is due to the private actions of their employers rather than the legislative regime itself. The former actions being unreviewable by virtue of this Court’s decision in RWDSU v. Dolphin Delivery Ltd., 1986 CanLII 5 (SCC), [1986] 2 S.C.R. 573, Sharpe J. dismissed the s. 2(d) claim. In response to the appellants’ claim that the LRESLAA constituted independently reviewable state action, Sharpe J. held that reviewing the LRESLAA would essentially constitutionalize the statute it repealed, namely, the ALRA. This would create “a broad class of statutes that would enjoy the status of a constitutional guarantee as they would be immune from repeal” (p. 208), an outcome rejected by the Ontario Court (General Division) in Ferrell v. Ontario (Attorney General) (1997), 1997 CanLII 12351 (ON SC), 149 D.L.R. (4th) 335.

9 Sharpe J. then turned to the s. 15(1) claim, which he felt was a more appropriate forum for addressing questions of positive state obligation under the Charter. With respect to the first prong of the s. 15(1) analysis, Sharpe J. held that agricultural workers had indeed been denied a legal benefit or protection enjoyed by most other workers, namely, the right to engage in statutory collective bargaining. However, Sharpe J. declined to recognize agricultural workers as an analogous group for the purpose of establishing discrimination under s. 15(1). In his view, the analogous grounds concept was rooted in the denial of human dignity and as such required the appellants to identify a “personal trait or characteristic” on which their differential treatment was based. It would not be sufficient, he held, to identify “occupational status” as such a characteristic, nor to combine occupational status with economic disadvantage. Thus, Sharpe J. rejected the s. 15(1) claim and with it the appellants’ constitutional challenge.

B. Ontario Court of Appeal (1999), [1999 CanLII 18653 (ON CA)](https://www.canlii.org/en/on/onca/doc/1999/1999canlii18653/1999canlii18653.html), 182 D.L.R. (4th) 471

10 Krever J.A., concurred in by Doherty and Rosenberg JJ.A., upheld the decision of Sharpe J., holding that “[w]e agree with the judgment of Sharpe J., both with the result at which he arrived and his reasons”.

V. Constitutional Questions

11 On June 20, 2000, Binnie J. stated the constitutional questions as follows:

1. Does s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Canadian Charter of Rights and Freedoms; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

2. Does s. 3(b) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Charter; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

3. If the answer to any part of questions 1 or 2 is in the affirmative, is the limitation nevertheless justified under s. 1 of the Charter?

VI. Analysis

A. Freedom of Association

(1) Nature of the Claim

12 The appellants claim that “[f]irst and foremost, agricultural workers simply wish to unionize”. Although the intervener Canadian Labour Congress raised the issue of collective bargaining in this appeal, the appellants directed this Court’s attention to broader issues, describing a range of union activities not related to collective bargaining. These activities, which ultimately relate to workers’ “empowerment and participation in both the workplace and society at large”, include promoting workplace democracy, protecting employees from abuses of managerial power, pooling resources, and expressing the views of workers “cogently and forcefully”. The appellants also described several social and political functions of trade unions, such as giving workers access to courts, bringing constitutional challenges on behalf of workers and engaging in political education and action. In my view, these functions make it clear that the appellants direct their attack not at legislation restricting collective bargaining per se, but at legislation restricting the “wider ambit of union purposes and activities”.

13 In order to establish a violation of s. 2(d), the appellants must demonstrate, first, that such activities fall within the range of activities protected by s. 2(d) of the Charter, and second, that the impugned legislation has, either in purpose or effect, interfered with these activities (see, in the s. 2(a) context, R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295, at pp. 331-36, and in the s. 2(b) context, Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927, at p. 971). On the first point, I shall review the existing framework of the s. 2(d) protection established by this Court in the 1987 “labour trilogy” and subsequent cases. This discussion will include a purposive analysis of s. 2(d), one which aims to protect the full range of associational activity contemplated by the Charter and to honour Canada’s obligations under international human rights law. After reviewing the content of freedom of association, I shall examine the contours of state responsibility under s. 2(d) of the Charter. In particular, I shall ask whether s. 2(d) obligates the state simply to respect trade union freedoms, or additionally to protect trade union freedoms by prohibiting their infringement by private actors. Following my discussion of the scope of s. 2(d), I shall examine the purpose and effects of the impugned legislation.

(2) Scope of Section 2(d)

(a) General Framework

14 The scope of s. 2(d) was first decided by this Court in a landmark trilogy of labour cases, all of which concerned the right to strike (see Reference re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313 (“Alberta Reference”); PSAC v. Canada, 1987 CanLII 89 (SCC), [1987] 1 S.C.R. 424; RWDSU v. Saskatchewan, 1987 CanLII 90 (SCC), [1987] 1 S.C.R. 460). In the Alberta Reference, McIntyre J. (writing for himself) stressed the double-edged nature of freedom of association, holding that “while [freedom of association] advances many group interests and, of course, cannot be exercised alone, it is nonetheless a freedom belonging to the individual and not to the group formed through its exercise” (p. 397). On the basis of this principle, McIntyre J. confined s. 2(d) to three elements: (1) the freedom to join with others in lawful, common pursuits and to establish and maintain organizations and associations (with which all six justices agreed), (2) the freedom to engage collectively in those activities which are constitutionally protected for each individual (with which three of six justices agreed) and (3) the freedom to pursue with others whatever action an individual can lawfully pursue as an individual (with which three of six justices agreed). These three elements of freedom of association are summarized, along with a crucial fourth principle, in the oft-quoted words of Sopinka J. in Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), 1990 CanLII 72 (SCC), [1990] 2 S.C.R. 367 (“PIPSC”), at pp. 401-2:

Upon considering the various judgments in the Alberta Reference, I have come to the view that four separate propositions concerning the coverage of the s. 2(d) guarantee of freedom of association emerge from the case: first, that s. 2(d) protects the freedom to establish, belong to and maintain an association; second, that s. 2(d) does not protect an activity solely on the ground that the activity is a foundational or essential purpose of an association; third, that s. 2(d) protects the exercise in association of the constitutional rights and freedoms of individuals; and fourth, that s. 2(d) protects the exercise in association of the lawful rights of individuals. [Emphasis added.]

The third and fourth of these principles have received considerably less judicial support than the others, having only been explicitly affirmed by three of six judges in the Alberta Reference and two of seven judges in PIPSC. Moreover, these elements of s. 2(d) provided little assistance to this Court in Canadian Egg Marketing Agency v. Richardson, 1997 CanLII 17020 (SCC), [1998] 3 S.C.R. 157 (“Egg Marketing”), which involved an activity that could not conceivably be performed by an individual. Most recently, in Delisle, supra, this Court did not have to rule on the validity of the existing framework because all of the activities involved fell within it. In that case, this Court clarified that s. 2(d) does not guarantee access to a particular labour relations regime where the claimants are able to exercise their s. 2(d) rights independently.

15 In addition to the four-part formulation in PIPSC, supra, an enduring source of insight into the content of s. 2(d) is the purpose of the provision. This purpose was first articulated in the labour trilogy and has accordingly been used to define both the “positive” freedom to associate as well as the “negative” freedom not to (see Alberta Reference, supra; Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211, at p. 318; R. v. Advance Cutting & Coring Ltd., [2001] 3 S.C.R. 209, 2001 SCC 70). In defining this purpose, McIntyre J. stressed, in Alberta Reference, supra, at p. 395, the unique power of associations to accomplish the goals of individuals:

While freedom of association like most other fundamental rights has no single purpose or value, at its core rests a rather simple proposition: the attainment of individual goals, through the exercise of individual rights, is generally impossible without the aid and cooperation of others. “Man, as Aristotle observed, is a ‘social animal, formed by nature for living with others’, associating with his fellows both to satisfy his desire for social intercourse and to realize common purposes.” (L. J. MacFarlane, The Theory and Practice of Human Rights (1985), p. 82.)

This conception of freedom of association, which was supported by Dickson C.J. in his dissenting judgment (at pp. 334 and 365-66), has been repeatedly endorsed by this Court since the Alberta Reference (see PIPSC, supra, per Sopinka J., at pp. 401-2, per Cory J. (dissenting), at p. 379; R. v. Skinner, 1990 CanLII 107 (SCC), [1990] 1 S.C.R. 1235, per Dickson C.J., at p. 1243; Lavigne, supra, per La Forest J., at p. 317, per Wilson J., at p. 251; per McLachlin J. (as she then was), at p. 343). In Lavigne, Wilson J. (writing for three of seven judges on this point) conducted an extensive review of this Court’s s. 2(d) jurisprudence, concluding that “this Court has been unanimous in finding on more than one occasion and in a variety of contexts that the purpose which s. 2(d) is meant to advance is the collective action of individuals in pursuit of their common goals” (p. 253). Wilson J. added that the Court has remained steadfast in this position despite numerous disagreements about the application of s. 2(d) to particular practices.

16 As these dicta illustrate, the purpose of s. 2(d) commands a single inquiry: has the state precluded activity because of its associational nature, thereby discouraging the collective pursuit of common goals? In my view, while the four-part test for freedom of association sheds light on this concept, it does not capture the full range of activities protected by s. 2(d). In particular, there will be occasions where a given activity does not fall within the third and fourth rules set forth by Sopinka J. in PIPSC, supra, but where the state has nevertheless prohibited that activity solely because of its associational nature. These occasions will involve activities which (1) are not protected under any other constitutional freedom, and (2) cannot, for one reason or another, be understood as the lawful activities of individuals. As discussed by Dickson C.J. in the Alberta Reference, supra, such activities may be collective in nature, in that they cannot be performed by individuals acting alone. The prohibition of such activities must surely, in some cases, be a violation of s. 2(d) (at p. 367):

There will, however, be occasions when no analogy involving individuals can be found for associational activity, or when a comparison between groups and individuals fails to capture the essence of a possible violation of associational rights. . . . The overarching consideration remains whether a legislative enactment or administrative action interferes with the freedom of persons to join and act with others in common pursuits. The legislative purpose which will render legislation invalid is the attempt to preclude associational conduct because of its concerted or associational nature. [Emphasis added.]

This passage, which was not explicitly rejected by the majority in the Alberta Reference or in PIPSC, recognizes that the collective is “qualitatively” distinct from the individual: individuals associate not simply because there is strength in numbers, but because communities can embody objectives that individuals cannot. For example, a “majority view” cannot be expressed by a lone individual, but a group of individuals can form a constituency and distill their views into a single platform. Indeed, this is the essential purpose of joining a political party, participating in a class action or certifying a trade union. To limit s. 2(d) to activities that are performable by individuals would, in my view, render futile these fundamental initiatives. At best, it would encourage s. 2(d) claimants to contrive individual analogs for inherently associational activities, a process which this Court clearly resisted in the labour trilogy, in Egg Marketing, supra, and in its jurisprudence on union security clauses and the right not to associate (see Syndicat catholique des employés de magasins de Québec Inc. v. Compagnie Paquet Ltée, 1959 CanLII 51 (SCC), [1959] S.C.R. 206 (“[t]he union is . . . the representative of all the employees in the unit for the purpose of negotiating the labour agreement”, hence “[t]here is no room left for private negotiation between employer and employee” (per Judson J., at p. 212)); McGavin Toastmaster Ltd. v. Ainscough, 1975 CanLII 9 (SCC), [1976] 1 S.C.R. 718 (“[t]he reality is, and has been for many years now throughout Canada, that individual relationships as between employer and employee have meaning only at the hiring stage” (per Laskin C.J., at p. 725)); I. Hunter, “Individual and Collective Rights in Canadian Labour Law” (1993), 22 Man. L.J. 145, at p. 147 (“[i]ndividual rights vis-à-vis their employer are replaced by rights in respect of their union, which, in turn, is mandated to advance the interests of bargaining-unit members”); D. Beatty and S. Kennett, “Striking Back: Fighting Words, Social Protest and Political Participation in Free and Democratic Societies” (1988), 67 Can. Bar Rev. 573, at pp. 587-88). The collective dimension of s. 2(d) is also consistent with developments in international human rights law, as indicated by the jurisprudence of the Committee of Experts on the Application of Conventions and Recommendations and the ILO Committee on Freedom of Association (see, e.g., International Labour Office, Freedom of Association: Digest of decisions and principles of the Freedom of Association Committee of the Governing Body of the ILO (4th ed. 1996)). Not only does this jurisprudence illustrate the range of activities that may be exercised by a collectivity of employees, but the International Labour Organization has repeatedly interpreted the right to organize as a collective right (see International Labour Office, Voices for Freedom of Association (Labour Education 1998/3, No. 112): “freedom is not only a human right; it is also, in the present circumstances, a collective right, a public right of organisation” (address delivered by Mr. Léon Jouhaux, workers’ delegate)).

17 As I see it, the very notion of “association” recognizes the qualitative differences between individuals and collectivities. It recognizes that the press differs qualitatively from the journalist, the language community from the language speaker, the union from the worker. In all cases, the community assumes a life of its own and develops needs and priorities that differ from those of its individual members. Thus, for example, a language community cannot be nurtured if the law protects only the individual’s right to speak (see R. v. Beaulac, 1999 CanLII 684 (SCC), [1999] 1 S.C.R. 768, at para. 20). Similar reasoning applies, albeit in a limited fashion, to the freedom to organize: because trade unions develop needs and priorities that are distinct from those of their members individually, they cannot function if the law protects exclusively what might be “the lawful activities of individuals”. Rather, the law must recognize that certain union activities -- making collective representations to an employer, adopting a majority political platform, federating with other unions -- may be central to freedom of association even though they are inconceivable on the individual level. This is not to say that all such activities are protected by s. 2(d), nor that all collectivities are worthy of constitutional protection; indeed, this Court has repeatedly excluded the right to strike and collectively bargain from the protected ambit of s. 2(d) (see Alberta Reference, supra, per Le Dain J., at p. 390 (excluding the right to strike and collectively bargain), per McIntyre J., at pp. 409-10 (excluding the right to strike); PIPSC, supra, per Dickson C.J., at pp. 373-74 (excluding the right to collectively bargain), per La Forest J., at p. 390 (concurring with Sopinka J.), per L’Heureux-Dubé J., at p. 392 (excluding both the right to strike and collectively bargain), per Sopinka J., at p. 404 (excluding both the right to strike and collectively bargain)). It is to say, simply, that certain collective activities must be recognized if the freedom to form and maintain an association is to have any meaning. As one author puts it, the per se exclusion of collective action reduces employee collectives to mere “aggregate[s] of economically self-interested individuals” rather than “co-operative undertakings where individual flourishing can be encouraged through membership in and co-operation with the community of fellow workers” (see L. Harmer, “The Right to Strike: Charter Implications and Interpretations” (1988), 47 U.T. Fac. L. Rev. 420, at pp. 434-35). This would surely undermine the purpose of s. 2(d), which is to allow the achievement of individual potential through interpersonal relationships and collective action (see, e.g., Lavigne, supra, per McLachlin J., at pp. 343-44, per La Forest J., at pp. 327-28).

18 In sum, a purposive approach to s. 2(d) demands that we “distinguish between the associational aspect of the activity and the activity itself”, a process mandated by this Court in the Alberta Reference (see Egg Marketing, supra, per Iacobucci and Bastarache JJ., at para. 111). Such an approach begins with the existing framework established in that case, which enables a claimant to show that a group activity is permitted for individuals in order to establish that its regulation targets the association per se (see Alberta Reference, supra, per Dickson C.J., at p. 367). Where this burden cannot be met, however, it may still be open to a claimant to show, by direct evidence or inference, that the legislature has targeted associational conduct because of its concerted or associational nature.

(b) State Responsibility Under Section 2(d)

19 The content of the freedom to organize having been discussed, the next question that arises is the scope of state responsibility in respect of this freedom. This responsibility is generally characterized as “negative” in nature, meaning that Parliament and the provincial legislatures need only refrain from interfering (either in purpose or effect) with protected associational activity. Conversely, the Charter does not oblige the state to take affirmative action to safeguard or facilitate the exercise of fundamental freedoms.

20 However, history has shown, and Canada’s legislatures have uniformly recognized, that a posture of government restraint in the area of labour relations will expose most workers not only to a range of unfair labour practices, but potentially to legal liability under common law inhibitions on combinations and restraints of trade. Knowing this would foreclose the effective exercise of the freedom to organize, Ontario has provided a statutory freedom to organize in its LRA (s. 5), as well as protections against denial of access to property (s. 13), employer interference with trade union activity (s. 70), discrimination against trade unionists (s. 72), intimidation and coercion (s. 76), alteration of working conditions during the certification process (s. 86), coercion of witnesses (s. 87), and removal of Board notices (s. 88). In this context, it must be asked whether, in order to make the freedom to organize meaningful, s. 2(d) of the Charter imposes a positive obligation on the state to extend protective legislation to unprotected groups. More broadly, it may be asked whether the distinction between positive and negative state obligations ought to be nuanced in the context of labour relations, in the sense that excluding agricultural workers from a protective regime substantially contributes to the violation of protected freedoms.

21 This precise question was raised in Delisle, supra, in which the appellant failed to establish that exclusion from a protective regime violated s. 2(d). The Delisle case involved RCMP officers who were employed by the Canadian government, so it is arguable that the Court’s decision was not intended to apply where private employers are involved. However, Justice L’Heureux‑Dubé recognized at para. 7 of a concurring judgment that s. 2(d) may require protection against unfair labour practices in certain circumstances:

I recognize that in cases where the employer does not form part of government, there exists no Charter protection against employer interference. In such a case, it might be demonstrated that the selective exclusion of a group of workers from statutory unfair labour practice protections has the purpose or effect of encouraging private employers to interfere with employee associations. It may also be that there is a positive obligation on the part of governments to provide legislative protection against unfair labour practices or some form of official recognition under labour legislation, because of the inherent vulnerability of employees to pressure from management, and the private power of employers, when left unchecked, to interfere with the formation and administration of unions. [Emphasis added.]

This dictum was not rejected by the Delisle majority, which focused instead on the fact that an interference with associational activity had not been made out on the facts of the case. Indeed, in making this finding, I deferred judgment on the appellant’s argument that underinclusion could have “an important chill on freedom of association because it clearly indicates to its members that unlike all other employees, they cannot unionize, and what is more, that they must not get together to defend their interests with respect to labour relations” (see Delisle, supra, at para. 30). In addition, I left open the possibility that s. 2 of the Charter may impose “a positive obligation of protection or inclusion on Parliament or the government . . . in exceptional circumstances which are not at issue in the instant case” (para. 33).

22 Even before Delisle, Le Dain J. recognized in the Alberta Reference, supra, that s. 2(d) protected workers’ freedom to organize “without penalty or reprisal”, making no distinction between workers employed by government or private entities (p. 391). What this dictum recognized, in my view, is that without the necessary protection, the freedom to organize could amount “to no more than the freedom to suffer serious adverse legal and economic consequences” (see H. W. Arthurs et al., Labour Law and Industrial Relations in Canada (4th ed. 1993), at para. 431). Perhaps more importantly for this appeal, this dictum implies that total exclusion from a regime protecting the freedom to organize could engage not only s. 15(1) of the Charter, but also s. 2(d) of the Charter. Where a group is denied a statutory benefit accorded to others, as is the case in this appeal, the normal course is to review this denial under s. 15(1) of the Charter, not s. 2(d) (see Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995; Native Women’s Assn. of Canada v. Canada, 1994 CanLII 27 (SCC), [1994] 3 S.C.R. 627 (“NWAC”); Delisle, supra). This was properly recognized by Sharpe J. who noted that “by ‘dipping its toe in the water’, and affording or enhancing the rights of some”, the government is not obliged to “go all the way and ensure the enjoyment of rights by all” (p. 207). However, it seems to me that apart from any consideration of a claimant’s dignity interest, exclusion from a protective regime may in some contexts amount to an affirmative interference with the effective exercise of a protected freedom. In such a case, it is not so much the differential treatment that is at issue, but the fact that the government is creating conditions which in effect substantially interfere with the exercise of a constitutional right; it has been held in the s. 2(a) context, for example, that “protection of one religion and the concomitant non-protection of others imports disparate impact destructive of the religious freedom of the collectivity” (see Big M Drug Mart, supra, at p. 337). This does not mean that there is a constitutional right to protective legislation per se; it means legislation that is underinclusive may, in unique contexts, substantially impact the exercise of a constitutional freedom.

23 This brings me to the central question of this appeal: can excluding agricultural workers from a statutory labour relations regime, without expressly or intentionally prohibiting association, constitute a substantial interference with freedom of association? A preliminary answer to this question may be found in Haig, supra, where L’Heureux‑Dubé J. recognized 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” (p. 1039). Although such a situation did not arise in that case, at least three observations are in order. First, the benefit sought in Haig, namely, participation in a national referendum, was, unlike inclusion in the LRA, not designed to safeguard the exercise of a fundamental freedom; thus, this Court was able to reject the appellants’ claim for positive state action on the grounds that it would constitutionalize a very limited statutory regime. Second, there was no evidence in Haig that without the benefit of the referendum, the appellant would have been incapable of expressing his views on Quebec secession; thus, the appellants failed to meet the minimum evidentiary burden required of a s. 2(b) claim (see Haig, at p. 1040). Finally, even had the appellant been unable to express his views on Quebec secession, that surely had nothing to do with his exclusion from the national referendum. Similar points may be made about NWAC, supra. In that case, this Court again recognized the possibility of positive government action in some cases, but concluded that the respondents’ exclusion from a particular series of constitutional discussions did not suppress their overall freedom of expression. As in the Haig case, the decisive point was the nature of the state action sought, combined with the absence of an evidentiary foundation for the s. 2 claim. By contrast, the appellants argue in this case that they possess no independent ability to organize, either inside or outside of the relevant statutory context.

24 In my view, the cases of Haig, NWAC and Delisle function to circumscribe, but not to foreclose, the possibility of challenging underinclusion under s. 2 of the Charter. One limit imposed by these cases is that claims of underinclusion should be grounded in fundamental Charter freedoms rather than in access to a particular statutory regime. Thus, in Haig, the majority of this Court held that “[a] government is under no constitutional obligation to extend [a referendum] to anyone, let alone to everyone”^*, and further that “[a] referendum as a platform of expression is . . . a matter of legislative policy and not of constitutional law” (p. 1041 (emphasis in original)). Similarly, in NWAC, the majority of this Court held that “[i]t cannot be claimed that NWAC has a constitutional right to receive government funding aimed at promoting participation in the constitutional conferences” (p. 654). In my view, the appellants in this case do not claim a constitutional right to general inclusion in the LRA, but simply a constitutional freedom to organize a trade association. This freedom to organize exists independently of any statutory enactment, even though the so-called “modern rights to bargain collectively and to strike” have been characterized otherwise in the Alberta Reference, supra, per Le Dain J., at p. 391. While it may be that the effective exercise of this freedom requires legislative protection in some cases, this ought not change the fundamentally non-statutory character of the freedom itself. As long as the appellants can plausibly ground their action in a fundamental Charter freedom, Haig and NWAC ought simply to be distinguished.

25 Second, the underinclusion cases demonstrate that a proper evidentiary foundation must be provided before creating a positive obligation under the Charter. This requirement proved fatal in Haig, NWAC and Delisle because the claimants in all three cases were unable to prove that the fundamental freedom at issue, as opposed to merely their requested statutory entitlement, was impossible to exercise. On the contrary, it was concluded in Haig that “the referendum itself, far from stifling expression, provided a particular forum for such expression” (p. 1040). Similarly, it was concluded in NWAC that “[e]ven assuming that in certain extreme circumstances, the provision of a platform of expression to one group may infringe the expression of another and thereby require the Government to provide an equal opportunity for the expression of that group, there was no evidence in this case to suggest that the funding or consultation of the four Aboriginal groups infringed the respondents’ equal right of freedom of expression” (p. 664). Finally, it was concluded in Delisle that “it is difficult to argue that the exclusion of RCMP members from the statutory regime of the PSSRA prevents the establishment of an independent employee association because RCMP members have in fact formed such an association in several provinces, including Quebec, where ‘C’ Division was created by Mr. Delisle himself” (para. 31). In my view, the evidentiary burden in these cases is to demonstrate that exclusion from a statutory regime permits a substantial interference with the exercise of protected s. 2(d) activity. Such a burden was implied by Dickson C.J. in the Alberta Reference, supra, where he stated that positive obligations may be required “where the absence of government intervention may in effect substantially impede the enjoyment of fundamental freedoms” (p. 361 (emphasis added)). It was also implied by this Court in NWAC, where Sopinka J. stated that “[i]t will be rare indeed that the provision of a platform or funding to one or several organizations will have the effect of suppressing another’s freedom of speech” (p. 657 (emphasis added)). These dicta do not require that the exercise of a fundamental freedom be impossible, but they do require that the claimant seek more than a particular channel for exercising his or her fundamental freedoms.

26 Assuming an evidentiary foundation can be provided, a third concern is whether the state can truly be held accountable for any inability to exercise a fundamental freedom. In this case, it is said that the inability to form an association is the result of private action and that mandating inclusion in a statutory regime would run counter to this Court’s decision in Dolphin Delivery, supra. However, it should be noted that this Court’s understanding of “state action” has matured since the Dolphin Delivery case and may mature further in light of evolving Charter values. For example, this Court has repeatedly held that the contribution of private actors to a violation of fundamental freedoms does not immunize the state from Charter review; rather, such contributions should be considered part of the factual context in which legislation is reviewed (see Lavigne, per La Forest J., at p. 309; see, similarly, R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713, per Dickson C.J., at p. 766). Moreover, this Court has repeatedly held in the s. 15(1) context that the Charter may oblige the state to extend underinclusive statutes to the extent underinclusion licenses private actors to violate basic rights and freedoms (see Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493). Finally, there has been some suggestion that the Charter should apply to legislation which “permits” private actors to interfere with protected s. 2 activity, as in some contexts mere permission may function to encourage or support the act which is called into question (see Lavigne, per Wilson J., at p. 248). If we apply these general principles to s. 2(d), it is not a quantum leap to suggest that a failure to include someone in a protective regime may affirmatively permit restraints on the activity the regime is designed to protect. The rationale behind this is that underinclusive state action falls into suspicion not simply to the extent it discriminates against an unprotected class, but to the extent it substantially orchestrates, encourages or sustains the violation of fundamental freedoms.

27 The notion that underinclusion can infringe freedom of association is not only implied by Canadian Charter jurisprudence, but is also consistent with international human rights law. Article 2 of Convention (No. 87) concerning Freedom of Association and Protection of the Right to Organize, 67 U.N.T.S. 17, provides that “[w]orkers and employers, without distinction whatsoever, shall have the right to establish and . . . to join organisations of their own choosing” (emphasis added), and that only members of the armed forces and the police may be excluded (Article 9). In addition, Article 10 of Convention No. 87 defines an “organisation” as “any organisation of workers or of employers for furthering and defending the interests of workers or of employers” (emphasis added). Canada ratified Convention No. 87 in 1972. The Convention’s broadly worded provisions confirm precisely what I have discussed above, which is that discriminatory treatment implicates not only an excluded group’s dignity interest, but also its basic freedom of association. This is further confirmed by the fact that Article 2 operates not only on the basis of sex, race, nationality and other traditional grounds of discrimination, but on the basis of any distinction, including occupational status (see L. Swepston, “Human rights law and freedom of association: Development through ILO supervision” (1998), 137 Int’l Lab. Rev. 169, at pp. 179-180). Nowhere is this clearer than in Article 1 of Convention (No. 11) concerning the Rights of Association and Combination of Agricultural Workers, 38 U.N.T.S. 153, which obliges ratifying member states to secure to “all those engaged in agriculture” the same rights of association as to industrial workers; the convention makes no distinction as to the type of agricultural work performed. Although provincial jurisdiction has prevented Canada from ratifying Convention No. 11, together these conventions provide a normative foundation for prohibiting any form of discrimination in the protection of trade union freedoms (see J. Hodges-Aeberhard, “The right to organise in Article 2 of Convention No. 87: What is meant by workers ‘without distinction whatsoever’?” (1989), 128 Int’l Lab. Rev. 177). This foundation is fortified by Convention (No. 141) concerning Organisations of Rural Workers and Their Role in Economic and Social Development (I.L.O. Official Bulletin, vol. LVIII, 1975, Series A, No. 1, p. 28) which extends, under Article 2, the freedom to organize to “any person engaged in agriculture, handicrafts or a related occupation in a rural area, whether as a wage earner or, . . . as a tenant, sharecropper or small owner-occupier”.

28 In sum, while it is generally desirable to confine claims of underinclusion to s. 15(1), it will not be appropriate to do so where the underinclusion results in the effective denial of a fundamental freedom such as the right of association itself. This is not to say that such claims will be common: they are constrained by both s. 32 of the Charter, which demands a minimum of state action before the Charter can be invoked, as well as by the factors discussed above. However, a claim for inclusion should not, in my view, automatically fail a s. 2(d) analysis: depending on the circumstances, freedom of association may, for example, prohibit the selective exclusion of a group from whatever protections are necessary to form and maintain an association, even though there is no constitutional right to such statutory protection per se. In this sense, the burden imposed by s. 2(d) of the Charter differs from that imposed by s. 15(1): while the latter focuses on the effects of underinclusion on human dignity (Law v. Canada (Minister of Employment and Immigration), 1999 CanLII 675 (SCC), [1999] 1 S.C.R. 497), the former focuses on the effects of underinclusion on the ability to exercise a fundamental freedom. This distinction is contemplated by the wording of the Charter itself and is supported by subsequent jurisprudence of this Court (see, e.g., Delisle, supra, at para. 25).

29 Before concluding on this point, I reiterate that the above doctrine does not, on its own, oblige the state to act where it has not already legislated in respect of a certain area. One must always guard against reviewing legislative silence, particularly where no legislation has been enacted in the first place. By the same token, it must be remembered why the Charter applies to legislation that is underinclusive. Once the state has chosen to regulate a private relationship such as that between employer and employee, I believe it is unduly formalistic to consign that relationship to a “private sphere” that is impervious to Charter review. As Dean P. W. Hogg has stated, “[t]he effect of the governmental action restriction is that there is a private realm in which people are not obliged to subscribe to ‘state’ values, and into which constitutional norms do not intrude. The boundaries of that realm are marked, not by an a priori definition of what is ‘private’, but by the absence of statutory or other governmental intervention” (see Constitutional Law of Canada ( loose-leaf ed.), at p. 34-27). I am not prepared to say that the relationship between farmers and their employees falls within that boundary. If, by investigating the effects of a statute that regulates this sphere, this Court is imposing “positive” obligations on the state, that is only because such imposition is justified in the circumstances.

(c) Summary of Discussion on Section 2(d)

30 In my view, the activities for which the appellants seek protection fall squarely within the freedom to organize, that is, the freedom to collectively embody the interests of individual workers. Insofar as the appellants seek to establish and maintain an association of employees, there can be no question that their claim falls within the protected ambit of s. 2(d) of the Charter. Moreover, the effective exercise of these freedoms may require not only the exercise in association of the constitutional rights and freedoms (such as freedom of assembly) and lawful rights of individuals, but the exercise of certain collective activities, such as making majority representations to one’s employer. These activities are guaranteed by the purpose of s. 2(d), which is to promote the realization of individual potential through relations with others, and by international labour jurisprudence, which recognizes the inevitably collective nature of the freedom to organize. Finally, while inclusion in legislation designed to protect such freedoms will normally be the province of s. 15(1) of the Charter, claims for inclusion may, in rare cases, be cognizable under the fundamental freedoms. With this in mind, I turn to whether s. 3(b) of the LRA interferes with the appellants’ protected freedoms, either in purpose or effect.

(3) Application to the Ontario Legislation

(a) Purpose of the Exclusion

31 The appellants claim that their exclusion from the LRA was intended to infringe their freedom to organize and, as such, violates the Charter notwithstanding its actual effects (see Big M Drug Mart, supra, at pp. 331-33; Edwards Books, supra, per Dickson C.J., at p. 752). A similar allegation of colourable purpose was assessed in the recent case of Delisle, supra. In that case, s. 2 “employee” (e) of the Public Service Staff Relations Act, R.S.C. 1985, c. P-35, was held not to interfere with the unionization of RCMP officers, as the purpose of the provision was simply to withhold from RCMP officers any status or protection created by the Act itself. The majority rejected extrinsic evidence that the purpose of the Act was, in the words of the dissenting judges, “to maintain the inherent difficulty faced by RCMP members in attempting to associate together to confront management on more equal terms” (para. 88). In the case at bar, a similar analysis yields an ambiguous result. At first blush, it would seem that the purpose of the LRA and the LRESLAA is to withhold from agricultural workers any status or protection created by the former Act, and not to target non-statutory unionization. On the other hand, the appellants point out several comments made by Ontario government officials to the effect that the purpose of the LRESLAA was to prevent “unionization”. Upon introducing the LRESLAA to the Ontario Legislature in 1995, for example, the Ontario Minister of Labour stated that “unionization of the family farm has no place in Ontario’s key agricultural sector”; moreover, the Minister of Agriculture, Food and Rural Affairs later stated that “the Agricultural Labour Relations Act is aimed directly at unionizing the family farm” and that “[w]e do not believe in the unionization of the family farm” (Legislative Assembly of Ontario, Official Report of Debates, October 4, 1995, at pp. 99-100). Similar language was employed in the legislature’s media kit on Bill 7, which stated that the agricultural sector “would have great difficulty adapting to the presence of unions”. These troubling comments were made to members of the provincial legislature before they voted on the LRESLAA and, as such, may have reflected the legislature’s intention in enacting that statute.

32 There are conflicting claims in this case concerning the meaning of the above comments and the light they shed on the intention of the legislature. On the one hand, the ambiguous use of the term “unionization” suggests that the legislature sought not only to exclude agricultural workers from the statutory incidents of striking and collective bargaining, but also to insulate Ontario’s farms from the very presence of unions. Such an intention would, needless to say, run counter to the Charter’s guarantee of freedom of association. On the other hand, the fact that the LRESLAA pursues a collateral legislative objective, namely the protection of the family farm, makes it difficult to conclude without speculation that this protection was sought through the prevention of unionization per se. While my colleague L’Heureux-Dubé J. marshals compelling evidence to make this point, I remain struck by the fact that s. 3(b) of the LRA does not, on its face, prohibit agricultural workers from forming workers’ associations, while it does bar them from all statutory labour relations schemes.

33 The difficulties of assessing legislative intent cannot be overemphasized. Such an assessment strikes at the heart of the rapport between the legislatures and the courts and, if undertaken lightly, can become a rather subjective process of induction. Moreover, the kind of evidence that is required to go behind the wording of a statute and make a finding of unconstitutional purpose is, understandably, not often available on the legislative record. On the facts of this case, therefore, I think it is more appropriate to focus on the effects of the impugned provisions, noting that some of the concerns raised by the above comments will inform the s. 1 analysis.

(b) Effects of the Exclusion

34 In their submissions before this Court, the appellants urged that because the statutory protections provided by the LRA were a necessary pre-condition for the formation of agricultural unions in Ontario, the effect of s. 3(b) of the LRA was to permanently foreclose this possibility and thus to violate s. 2(d) of the Charter. In response, the Attorney General adopted the position of Sharpe J. that LRA protection was an insufficient condition for the formation of agricultural unions and, more importantly, that any inability to form agricultural unions in Ontario stemmed from private, not state action. In my view, the appellants must prevail on this point. While the respondent rightly observes that the Charter does not apply to private actors, their argument assumes a rigid dichotomy between public and private action which, while appropriate in some contexts, belies the historical reality of agricultural labour relations. I conclude that the effect of s. 3(b) of the LRA is to violate s. 2(d) of the Charter.

35 The history of labour relations in Canada illustrates the profound connection between legislative protection and the freedom to organize. It may be suggested that legislative protection is so tightly woven into the fabric of labour relations that, while there is no constitutional right to protective legislation per se, the selective exclusion of a group from such legislation may substantially impact the exercise of a fundamental freedom. To illustrate this point, I find it necessary to make three observations about the appellants’ exclusion from the LRA. First, the LRA is designed to safeguard the exercise of a fundamental freedom, rather than to provide a limited statutory entitlement to certain classes of citizens. Second, the appellants in this case are substantially incapable of exercising their fundamental freedom to organize without the protective regime, as indicated by the record filed before this Court. Third, the appellants’ exclusion from the LRA functions not simply to permit private interference with their fundamental freedoms, but to substantially reinforce such interferences. Central to all of these points, in my view, is that the freedom to organize constitutes a unique swatch in Canada’s constitutional fabric, as difficult to exercise as it is fundamental, into which legislative protection is historically woven.

(i) The [LRA](https://www.canlii.org/en/on/laws/stat/so-1995-c-1-sch-a/latest/so-1995-c-1-sch-a.html) is Designed to Safeguard the Exercise of the Fundamental Freedom to Associate

36 In assessing the appellant’s claim for the repeal of s. 3(b) of the LRA, it is crucial to examine the essential ambition of the LRA. As numerous scholars have pointed out, the LRA does not simply enhance, but instantiates, the freedom to organize. The Act provides the only statutory vehicle by which employees in Ontario can associate to defend their interests and, moreover, recognizes that such association is, in many cases, otherwise impossible. This recognition is evident not only from the statute’s protections against unfair labour practices, but from the express “right to organize” it inscribes in s. 5. At the same time, the activities for which the appellants seek protection antecede, at least notionally, the LRA’s enactment; as this Court held in Delisle, supra, “[t]he ability to form an independent association and to carry on [its] protected activities . . . exists independently of any statutory regime”, even though the unprotected aspects of collective bargaining and the right to strike are creatures of statute (para. 33). What this means is that, while the inevitable effect of allowing this appeal may be to extend a statutory regime to agricultural workers, depending on the legislative response to this decision, the appellants are not seeking a constitutional “right” to inclusion in the LRA.

37 The freedom to organize lies at the core of the Charter’s protection of freedom of association. So central is this freedom to s. 2(d) that, during the legislative hearings preceding the Charter’s enactment, an express right to unionize was opposed on the grounds “that that is already covered in the freedom of association that is provided already in . . . the Charter” (emphasis added) (see Minutes of Proceedings and Evidence of the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada, Issue No. 43, January 22, 1981, at pp. 69-70 (Kaplan)). As recently as Delisle, supra, L’Heureux‑Dubé J. noted that “the right to freedom of association must take into account the nature and importance of labour associations as institutions that work for the betterment of working conditions and the protection of the dignity and collective interests of workers in a fundamental aspect of their lives: employment” (para. 6 (emphasis in original)). These remarks echo those of Dickson C.J., who noted in the Alberta Reference, supra, that “[w]ork is one of the most fundamental aspects in a person’s life, providing the individual with a means of financial support and, as importantly, a contributory role in society” (p. 368) (see similarly, McKinney v. University of Guelph, 1990 CanLII 60 (SCC), [1990] 3 S.C.R. 229, per La Forest J., at p. 300; Machtinger v. HOJ Industries Ltd., 1992 CanLII 102 (SCC), [1992] 1 S.C.R. 986, at p. 1002; Wallace v. United Grain Growers Ltd., 1997 CanLII 332 (SCC), [1997] 3 S.C.R. 701, per Iacobucci J., at para. 95). Moreover, the importance of trade union freedoms is widely recognized in international covenants, as is the freedom to work generally. In my view, judicial recognition of these freedoms strengthens the case for their positive protection. It suggests that trade union freedoms lie at the core of the Charter, and in turn that legislation instantiating those freedoms ought not be selectively withheld where it is most needed.

38 By protecting the freedom to organize, s. 2(d) of the Charter recognizes the dynamic and evolving role of the trade union in Canadian society. In addition to permitting the collective expression of employee interests, trade unions contribute to political debate. At the level of national policy, unions advocate on behalf of disadvantaged groups and present views on fair industrial policy. These functions, when viewed globally, affect all levels of society and constitute “an important subsystem in a democratic market-economy system” (see K. Sugeno, “Unions as social institutions in democratic market economies” (1994), 133 Int’l Lab. Rev. 511, at p. 519). For these reasons, the notion that minimum legislative protection cannot be extended to agricultural workers without extending full collective bargaining rights is misguided. Equally misguided is the notion that inherent difficulties in the formation of trade unions, or the fact that unions are in some cases experiencing a decline in membership, diminishes their social and political significance. On the contrary, unions remain core voluntary associations based on the principle of freedom of association.

(ii) Without the Protection of the LRA, Agricultural Workers Are Substantially Incapable of Exercising the Freedom to Associate

39 The fact that a regime aims to safeguard a fundamental freedom does not, of course, mean that exclusion from that regime automatically gives rise to a Charter violation. As I discussed in Delisle, supra, a group that proves capable of associating despite its exclusion from a protective regime will be unable to meet the evidentiary burden required of a Charter claim. In such a case, inclusion in a statutory regime cannot be said to safeguard, but rather to enhance, the exercise of a fundamental freedom. In this case, by contrast, the appellants contend that total exclusion from the LRA creates a situation whereby they are substantially incapable of exercising their constitutional right to associate. Needless to say, this claim must be assessed against the factual record provided by both the appellants and the respondents.

40 As a preliminary matter, the appellants state that the repeal of the ALRA by the LRESLAA caused the immediate demise of the first agricultural workers’ union in Ontario. While this is an alluring argument, in my view it obscures the true substance of the appellants’ claim. As discussed above, what is ultimately impugned in this case is not simply the repeal of the ALRA, but the combined effect of the LRESLAA and the LRA. This implicates the decades-long exclusion of agricultural workers from the labour relations regime, from the first enactment of the Collective Bargaining Act, 1943, until the repeal of the ALRA in 1995. The LRESLAA occupies only a small space in this history; it ought not prove decisive in this appeal.

41 Nonetheless, the appellants argue that notwithstanding the ALRA, they have no realistic chance of associating without the protection of the LRA. This is mainly because the LRA protects workers from common law inhibitions on organizing activity, as well as from employer practices designed to obstruct the formation of unions (see Arthurs, supra, at para. 431; International Labour Organization, Committee on Freedom of Association, Report No. 308, Case No. 1900, “Complaint against the Government of Canada (Ontario), presented by the Canadian Labour Congress (CLC)”, I.L.O. Official Bulletin, vol. LXXX, 1997, Series B, No. 3, at paras. 145-46 and 187). Perhaps more broadly, the LRA is described by the intervener Canadian Labour Congress as having “regulated, structured and channelled” the method through which Canadian workers are able to organize, to the point where organizing a workers’ association is “virtually synonymous” with unionizing under the legislative scheme. As just noted, the mere fact of exclusion from protective legislation is not conclusive evidence of a Charter violation; as I observed in Delisle, supra, RCMP officers had the strength to form employee associations in several provinces despite their exclusion from the PSSRA (para. 31). That being said, it is possible to draw a distinction between groups who are “strong enough to look after [their] interests without collective bargaining legislation” and those “who have no recourse to protect their interests aside from the right to quit” (see Canadian Industrial Relations: The Report of the Task Force on Labour Relations (1968), at paras. 253-54). As Canada’s leading Task Force on Labour Relations recognized as early as 1968, agricultural workers fall into the latter category (para. 254). Not only have agricultural workers proved unable to form employee associations in provinces which deny them protection but, unlike the RCMP officers in Delisle, they argue that their relative status and lack of statutory protection all but guarantee this result. Distinguishing features of agricultural workers are their political impotence, their lack of resources to associate without state protection and their vulnerability to reprisal by their employers; as noted by Sharpe J., agricultural workers are “poorly paid, face difficult working conditions, have low levels of skill and education, low status and limited employment mobility” (p. 216). Moreover, unlike RCMP officers, agricultural workers are not employed by the government and therefore cannot access the Charter directly to suppress an unfair labour practice (Delisle, at para. 32). It is no wonder, therefore, according to the appellants, that agricultural workers have failed to associate in any meaningful way in Ontario, while RCMP officers have successfully created independent employee associations in several provinces across Canada (Delisle, at para. 31).

42 The validity of this claim will depend in part on how strict a definition of the word “unionize” this Court adopts. The respondent Fleming Chicks and the intervener Labour Issues Coordinating Committee both adopt a very strict definition, arguing that UFCW’s involvement in this litigation proves that the LRA has not functioned to stifle union activity. This claim is disputed by the appellants’ chief expert, Professor Judy Fudge, who notes that legislative protection is a necessary precondition for collective bargaining under Canadian labour relations legislation. As stated earlier in these reasons, it is only the right to associate that is at issue here, not the right to collective bargaining. Nevertheless, to suggest that s. 2(d) of the Charter is respected where an association is reduced to claiming a right to unionize would, in my view, make a mockery of freedom of association. The record shows that, but for the brief period covered by the ALRA, there has never been an agricultural workers’ union in Ontario. Agricultural workers have suffered repeated attacks on their efforts to unionize. Conversely, in those provinces where labour relations rights have been extended to agricultural workers, union density is higher than in Ontario (see Statistics Canada, Annual Report of the Minister of Industry, Science and Technology under the Corporations and Labour Unions Returns Act, Part II, Labour Unions (1992), at pp. 38-41). The respondents do not contest this evidence, nor do they deny that legislative protection is absolutely crucial if agricultural workers wish to unionize. Indeed, to suggest otherwise would contradict a widespread consensus among Parliament and the provincial legislatures that without certain minimum protections, the somewhat limited freedom to organize itself would be a hollow freedom. For these reasons, I readily conclude that the evidentiary burden has been met in this case: the appellants have brought this litigation because there is no possibility for association as such without minimum statutory protection.

(iii) The Exclusion of Agricultural Workers from the LRA Substantially Reinforces the Inherent Difficulty in Exercising the Freedom to Associate

43 Their freedom to organize having been substantially impeded by exclusion from protective legislation, it is still incumbent on the appellants to link this impediment to state, not just private action (see Dolphin Delivery, supra). On this point, the respondents argue that since agricultural workers are isolated, seasonal and relatively under-educated, this, along with the unfair labour practices of their employers, is what explains the difficulty in creating associations rather than the underinclusiveness of the legislation. On the other hand, the appellants argue that the above conditions are reinforced by legislation which fails to provide minimum protection of their freedom to organize and further isolates agricultural workers by excluding them from the general regime of labour relations.

44 In my view, the appellants’ argument must prevail. What the legislature has done by reviving the LRA is not simply allow private circumstances to subsist; it has reinforced those circumstances by excluding agricultural workers from the only available channel for associational activity (see Vriend, supra, at paras. 99-103). The most poignant chapter in this legislative history, but by no means the decisive one, is the LRESLAA. Through this enactment, the Ontario government not only renewed its commitment to preventing agricultural unions from collective bargaining, but prohibited even the voluntary recognition of agricultural associations, whatever their attributes might be. At the same time, it must be presumed that the legislature understood the history of labour relations and remained of the view that a protective regime was essential to the exercise of freedom of association in this area.

45 The most palpable effect of the LRESLAA and the LRA is, in my view, to place a chilling effect on non-statutory union activity. By extending statutory protection to just about every class of worker in Ontario, the legislature has essentially discredited the organizing efforts of agricultural workers. This is especially true given the relative status of agricultural workers in Canadian society. In Delisle, supra, I linked RCMP officers’ ability to associate to their relative status, comparing them with the armed forces, senior executives in the public service and judges. The thrust of this argument was that if the PSSRA sought to discourage RCMP officers from associating, it could not do so in light of their relative status, their financial resources and their access to constitutional protection. By contrast, it is hard to imagine a more discouraging legislative provision than s. 3(b) of the LRA. The evidence is that the ability of agricultural workers to associate is only as great as their access to legal protection, and such protection exists neither in statutory nor constitutional form. Moreover, agricultural workers already possess a limited sense of entitlement as a result of their exclusion from other protective legislation related to employment standards and occupational health and safety (see Employment Standards Act Regulations, R.R.O. 1990, Reg. 325, s. 3(1)(i), excluding most agricultural workers from Parts IV-VIII of the Employment Standards Act, R.S.O. 1990, c. E.14; Occupational Health and Safety Act, R.S.O. 1990, c. O.1, s. 3(2)). In this context, the effect of s. 3(b) of the LRA is not simply to perpetuate an existing inability to organize, but to exert the precise chilling effect I declined to recognize in Delisle.

46 Conversely, the didactic effects of labour relations legislation on employers must not be underestimated. It is widely accepted that labour relations laws function not only to provide a forum for airing specific grievances, but for fostering dialogue in an otherwise adversarial workplace. As P. Weiler has written, unionization introduces a form of political democracy into the workplace, subjecting employer and employee alike to the “rule of law” (see Reconcilable Differences: New Directions in Canadian Labour Law (1980), at pp. 31-32). In this context, the wholesale exclusion of agricultural workers from a labour relations regime can only be viewed as a stimulus to interfere with organizing activity. The exclusion suggests that workplace democracy has no place in the agricultural sector and, moreover, that agricultural workers’ efforts to associate are illegitimate. As surely as LRA protection would foster the “rule of law” in a unionized workplace, exclusion from that protection privileges the will of management over that of the worker. Again, a contrast to Delisle, supra, is apposite: a government employer is less likely than a private employer to take exclusion from protective legislation as a green light to commit unfair labour practices, as its employees have direct recourse to the Charter.

47 For these reasons, I believe it is inappropriate for the Ontario Legislature to distance itself from the effects of the LRA and the LRESLAA. The enactment of the Collective Bargaining Act, 1943 reflected the legislature’s awareness of employer unfair labour practices and its concomitant recognition that legislation was necessary to enable workers’ freedom of association. The Collective Bargaining Act, 1943 was enacted against a background of staunch resistance to the labour movement; in large part, it was intended to prevent discrimination against union members. In this context, the exclusion of an entire category of workers from the LRA can only be viewed as a foreseeable infringement of their Charter rights. It was obviously open to the respondents to argue that the legislature has since altered its view of the need for protective legislation and that the LRA is not even required for the majority of workers today. However, by reviving the exclusion in 1995 and providing time-limited protection against penalty and reprisal, the legislature clearly acknowledged otherwise (see LRESLAA, s. 81(1)). In essence, after recognizing agricultural workers’ need for protection, the legislature made things more difficult for them by excluding them from the protective regime put in place in 1943. For these reasons, the respondents cannot claim that circumstances have changed substantially since the enactment of the Collective Bargaining Act, 1943; rather, it can only justify the exclusion of agricultural workers on the basis of collateral concerns such as the protection of the family farm and the need to maintain a competitive agricultural sector – issues which, needless to say, must be considered under s. 1 of the Charter.

48 In sum, I believe it is reasonable to conclude that the exclusion of agricultural workers from the LRA substantially interferes with their fundamental freedom to organize. The inherent difficulties of organizing farm workers, combined with the threats of economic reprisal from employers, form only part of the reason why association is all but impossible in the agricultural sector in Ontario. Equally important is the message sent by s. 3(b) of the LRA, which delegitimizes associational activity and thereby ensures its ultimate failure. Given these known and foreseeable effects of s. 3(b), I conclude that the provision infringes the freedom to organize and thus violates s. 2(d) of the Charter.

B. Section 1

49 Having established a violation of s. 2(d) of the Charter, the question arises as to whether exclusion from the LRA constitutes a reasonable limit on agricultural workers’ freedom to organize. In this regard, s. 1 of the Charter obliges the respondents, as the parties seeking to uphold the limitation, to establish both that the objective underlying the limitation is of sufficient importance to warrant overriding a constitutionally protected right or freedom, and that the means chosen to reach this objective are proportionate (see R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103, at pp. 136-39). This analysis must be undertaken with a close attention to the factual and social context surrounding the enactment of the LRA; as I noted in Thomson Newspapers Co. v. Canada (Attorney General), 1998 CanLII 829 (SCC), [1998] 1 S.C.R. 877, at para. 87, “context is the indispensable handmaiden to the proper characterization of the objective of the impugned provision, to determining whether that objective is justified, and to weighing whether the means used are sufficiently closely related to the valid objective so as to justify an infringement of a Charter right”. The contextual factors established in Thomson Newspapers and subsequent cases will prove especially helpful at the minimum impairment stage of the s. 1 analysis.

(1) Sufficiently Important Objective

50 According to settled s. 1 jurisprudence, the respondents must establish that the objectives of the infringing measures, in this case s. 80 of the LRESLAA and s. 3(b) of the LRA, “are of sufficient importance to warrant overriding a constitutionally protected right or freedom” (Big M Drug Mart, supra, at p. 352). The appellants argue that these objectives must be those “that originally motivated the government action in question”, and that “[t]he exclusion contained in the 1943 Act simply followed the approach taken in the U.S. Wagner Act which had excluded agricultural workers on racial grounds”. In support of this view, the appellants note that at the time the Wagner Act was enacted, agricultural workers in the American South were predominantly Black and, due to their alienation from the political process, were unable to prevail over the will of powerful Southern Democrats. In my view, this argument confuses the objective underlying the passage of the 1943 Act with the social and political factors surrounding its enactment. While it may be that Southern Democrats held the balance of power at the time the Wagner Act was enacted, and further that the majority of agricultural workers lacked access to the political process on account of their race, this does not prove that these workers were excluded “on racial grounds”. What it establishes, rather, is that the “administrative reasons” cited for excluding agricultural workers were accepted without debate because the workers themselves lacked an effective political voice. While this undoubtedly taints the legacy of the Wagner Act and the LRA, it does not alter the apparent policy objectives underlying the exclusion of agricultural workers.

51 The respondent Attorney General for Ontario proffers two broad objectives in this case, no doubt designed to avert the suggestion that its objectives have shifted since the 1943 Act:

(1) to recognize the unique characteristics of Ontario agriculture and its resulting incompatibility with legislated collective bargaining; and

(2) to further the purpose of the LRA by extending legislated collective bargaining only to fields of employment where the Act’s purposes can be realized.

While it is widely recognized that certain occupations may, in certain cases, be incompatible with collective bargaining, the judiciary and some essential services, for example, it is less certain that agricultural workers fall into this category. The Attorney General tenders extensive affidavit evidence on this point, arguing that the prevalence of the “family farm” and the vulnerability of the agricultural production process militate against legislated collective bargaining. For their part, the appellants maintain that the family farm no longer typifies Ontario agriculture and that the vulnerability of the agricultural production process, assuming it exists, does not militate against legislated collective bargaining. This discussion is however somewhat irrelevant in that the breach of the right of association does not extend to collective bargaining. What the government of Ontario must justify with regard to this appeal is its substantial interference with the right to form agricultural associations.

52 Judging from the parties’ evidence, I am satisfied both that many farms in Ontario are family-owned and -operated, and that the protection of the family farm is a pressing enough objective to warrant the infringement of s. 2(d) of the Charter. The fact that Ontario is moving increasingly towards corporate farming and agribusiness does not, in my view, diminish the importance of protecting the unique characteristics of the family farm; on the contrary, it may even augment it. Perhaps more importantly, the appellants do not deny that the protection of the family farm is, at least in theory, an admirable objective. The choice to “pursue the pastoral path” implies a unique and non-commercial way of life; this way of life is entitled, according to many, to the same level of protection as “that which prevails in our factories and office buildings” (D. M. Beatty, Putting the Charter to Work: Designing a Constitutional Labour Code (1987), at p. 91). If providing this protection means restraining the activities of those who would interfere with that choice, the appropriate response is not to deny the protection, however, but to balance these interests against one another. Such balancing, in my view, is the essence of s. 1 of the Charter (see R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2, at para. 97). As Professor Beatty puts it, at p. 91, “the freedom of those who choose to experience their lives in such non-commercial, self-sustaining ways may justify restraining the freedom of others who would wish to associate with them in a way which would threaten or deny them the opportunity to realize their choice”.

53 With respect to the economic rationale, I disagree with the appellants that “[t]he Government has provided no evidence that the Ontario agricultural sector is in a fragile competitive position or that it is likely to be substantially affected by small changes in the cost and operating structure of Ontario farming”. The Attorney General notes that agriculture occupies a volatile and highly competitive part of the private sector economy, that it experiences disproportionately thin profit margins and that its seasonal character makes it particularly vulnerable to strikes and lockouts. Moreover, these characteristics were readily accepted by the Task Force leading to the adoption of the ALRA, which recommended a system of compulsory arbitration in order to guard against the economic consequences of strikes and lockouts (see Task Force on Agricultural Labour Relations (Ontario), Second Report to the Minister of Labour (November 1992), at pp. 2 and 7). Whether such a recommendation is more constitutionally reasonable than a wholesale exclusion of agricultural workers is, in my view, an issue of proportionality rather than pressing and substantial objective. In other words, accepting the importance of protecting the family farm and ensuring farm productivity, the crucial question is whether the total exclusion of agricultural workers from the LRA is (1) a rational way of achieving this objective, (2) a measured response to this objective, and (3) not so severe in its effects that the Charter breach outweighs the objective’s importance (Oakes, supra).

(2) Proportionality

(a) Rational Connection

54 At this stage, the question is whether a wholesale exclusion of agricultural workers from the LRA is carefully tailored to meet its stated objectives. Put differently, can the formation of agricultural unions rationally be regarded as a threat to the unique characteristics of Ontario’s agriculture? Or conversely, does a regime which substantially impedes the right to form agricultural unions advance the cause articulated by the Attorney General? In my view, the Attorney General has demonstrated that unionization involving the right to collective bargaining and to strike can, in certain circumstances, function to antagonize the family farm dynamic. The reality of unionization is that it leads to formalized labour-management relationships and gives rise to a relatively formal process of negotiation and dispute resolution; indeed, this may well be its principal advantage over a system of informal industrial relations. In this context, it is reasonable to speculate that unionization will threaten the flexibility and cooperation that are characteristic of the family farm and distance parties who are otherwise, to use the respondent’s words, “interwoven into the fabric of private life” on the farm. That said, I hasten to add that this concern ought only be as great as the extent of the family farm structure in Ontario and that it does not necessarily apply to the right to form an agricultural association. In cases where the employment relationship is formalized to begin with, preserving “flexibility and co-operation” in the name of the family farm is not only irrational, it is highly coercive. The notion that employees should sacrifice their freedom to associate in order to maintain a flexible employment relationship should be carefully circumscribed, as it could, if left unchecked, justify restrictions on unionization in many sectors of the economy.

55 Even less convincing than Ontario’s family farm policy, in my view, is a policy of denying the right of association to agricultural workers on economic grounds. While this may be a rational policy in isolation, it is nothing short of arbitrary where collective bargaining rights have been extended to almost every other class of worker in Ontario. The reality, as acknowledged by all parties to this appeal, is that many industries experience thin profit margins and unstable production cycles; this may be due to unpredictable and time-sensitive weather conditions, as in the case of agriculture, or to other factors such as consumer demand and international competition. In my view, it would be highly arbitrary to accept this reasoning in respect of almost every industry in Ontario, only to extend it in respect of vulnerable agricultural workers to the point of denying them the right to associate. As Professor Beatty has written, “[i]f indeed collective bargaining increases the costs of labour to the overall detriment of society, then our legislators should repeal the legislation in its entirety rather than selectively exclude those most in need of its protection” (Beatty, supra, at p. 90). I conclude that the respondents have not met the onus of proof with regard to the economic rationale.

(b) Minimum Impairment

56 The next issue is whether recognizing the unique characteristics of Ontario agriculture and its resulting incompatibility with the formation of agricultural associations as a reasonable minimum justifies the complete exclusion of agricultural workers from the LRA. The LRA excludes all persons “employed in agriculture, hunting or trapping” from its application, defining agriculture as “farming in all its branches, including dairying, beekeeping, aquaculture, the raising of livestock including non-traditional livestock, furbearing animals and poultry, the production, cultivation, growing and harvesting of agricultural commodities, including eggs, maple products, mushrooms and tobacco, and any practices performed as an integral part of an agricultural operation” (see LRA, s. 1(1)). This provision has been broadly interpreted by Ontario’s Labour Relations Board, albeit with some reluctance and interpretive difficulty. In Wellington Mushroom Farm, [1980] O.L.R.B. Rep. May 813, for example, a majority of the board denied LRA certification to the employees of a mushroom factory, even though the actual growing of the mushrooms took place within a single-storey concrete block building. The majority of the board recognized that the employer’s operation did “not differ in any material respect from a typical manufacturing plant”, but it concluded that the growing of mushrooms constituted an agricultural activity in the ordinary sense of the term (p. 819). In other cases, the board has denied LRA protection to stationary engineers employed at a greenhouse, truck drivers hired to transport chickens and employees of a chicken hatchery (see Calvert-Dale Estates Ltd., [1971] O.L.R.B. Rep. Feb. 58; Spruceleigh Farms, [1972] O.L.R.B. Rep. Oct. 860; Cuddy Chicks Ltd., [1988] O.L.R.B. Rep. May 468, application for judicial review dismissed (1988), 1988 CanLII 4839 (ON SC), 66 O.R. (2d) 284 (Div. Ct.), aff’d (1989), 1989 CanLII 4139 (ON LRB), 70 O.R. (2d) 179 (C.A.), aff’d 1991 CanLII 57 (SCC), [1991] 2 S.C.R. 5; see also, G. W. Adams, Canadian Labour Law (2nd ed. (loose-leaf)), at pp. 6‑49 and 6-50).

57 The Attorney General claims, rightly in my view, that to exclude a given occupation from the LRA “involves a weighing of complex values and policy considerations that are often difficult to balance” and that this balancing “will in large part depend upon the particular perspective, priorities, views, and assumptions of the policy makers, as well as the political and economic theory to which they subscribe”. Similar statements have been made about labour relations generally, which have been described as “an extremely sensitive subject” premised on “a political and economic compromise between organized labour -- a very powerful socio-economic force -- on the one hand, and the employers of labour -- an equally powerful socio-economic force -- on the other” (Alberta Reference, supra, per McIntyre J., at p. 414). Policy choices are based on value judgments. This Court will only interfere with such choices where a more fundamental value is at stake and where it is apparent that a free and democratic society cannot permit the policy to interfere with the right in the circumstances of the case. The basis for the policy choice must be questioned strictly. It is not the motive of the legislature that is at issue, but the foundation for its policy. What is justified is that which is based on a general public purpose, is practically necessary and has a rational basis that can be supported after a normative evaluation of the area of intervention. Given the delicate balance between interests that is required here, as well as the added complexity of protecting the character of the family farm, one might be tempted to conclude that a wide margin of deference is owed to the enacting legislature when applying the minimum impairment test (see Thomson Newspapers, supra, at paras. 111-15). However, as outlined in Thomson Newspapers, political complexity is not the deciding factor in establishing a margin of deference under s. 1. Rather, the margin will vary according to whether legislature has (1) sought a balance between the interests of competing groups, (2) defended a vulnerable group with a subjective apprehension of harm, (3) chosen a remedy whose effectiveness cannot be measured scientifically, and (4) suppressed an activity whose social or moral value is relatively low. In my view, these factors on the whole favour a strict application of the minimum impairment test in the context of this appeal.

58 In Delisle, supra, Cory and Iacobucci JJ. applied the above factors to a provision much like the one impugned in this case, concluding that “none of the contextual factors discussed by Bastarache J. in Thomson Newspapers favours an exercise of deference to the legislature” (para. 128 (emphasis in original)). In my view, this analysis should be applied to this case with two minor exceptions. With respect to the second factor, the vulnerable group at issue in this case is a constituency of family farmers whose unique way of life stands to be jeopardized by collective bargaining legislation. This suggests that the margin of deference ought to be widened, although I hasten to add that the appellants also represent a vulnerable group worthy of legislative protection. The third factor yields a similarly ambiguous result. On the one hand, there is no concrete evidence to refute the conclusion that totally excluding agricultural workers from the LRA achieves the legislature’s objectives. Nor is there any scientific way of measuring whether the formation of agricultural unions undermines the family farm lifestyle. However, one can surely draw on statistical evidence, as both parties have in this case, to determine the prevalence of the family farm in Ontario. Given the centrality of this issue to the respondents’ case, it would be inappropriate to accord deference to the legislature and avoid evaluating their statistical claims.

59 With this in mind, I turn to the question of whether the legislature has impeded the appellants’ associational activity more than is reasonably necessary to achieve its stated objectives. The Attorney General makes three arguments in defence of its policy: first, that unionization is not appropriate for the “vast majority” of Ontario agricultural operations; second, that no appropriate dispute resolution mechanism exists for agricultural workers; and third, that extending collective bargaining rights to certain sectors of agriculture would be “arbitrary and impracticable”. It is also submitted that legislatures are entitled to a margin of deference when balancing complex matters of economic policy. In my view, these arguments all fail on a single point: they do not justify the categorical exclusion of agricultural workers where no satisfactory effort has been made to protect their basic right to form associations.

60 This effort need not produce the result most desirable to this Court; however, the legislature must “attempt very seriously to alleviate the effects” of its laws on those whose fundamental freedoms are infringed (Edwards Books, supra, per Dickson C.J., at p. 782). In Edwards Books, for example, this Court considered whether a Sunday closure law constituted a justifiable limit on the religious freedom of Saturday observers. The law provided an exemption for stores that had been closed on Saturday. However, the exemption was withheld from stores that employed more than seven employees and occupied 5,000 square feet of retail space or more. Despite the existence of less intrusive legislation in other provinces, three members of this Court characterized the exemption as “a satisfactory effort on the part of the Legislature of Ontario” (p. 782) and, on that basis, upheld the Act under s. 1. By contrast, neither enactment in this case includes a concrete attempt to alleviate the infringing effects on agricultural workers. Not only does the legislation fail to distinguish between different types of agriculture, but there is no evidence that the Ontario Legislature even turned its mind to freedom of association when enacting either statute. Rather, each enactment merely cloned a piece of existing legislation -- be it the Wagner Act or the LRA -- and, in so doing, relied on studies that had been commissioned by a previous legislature.

61 In my view, there are at least two ways in which the LRESLAA might impair Charter rights more than is reasonably necessary to achieve its objectives: first, by denying the right of association to every sector of agriculture, and second, by denying every aspect of the right, specially whatever protection is necessary to form and maintain employee associations, to agricultural workers. A similar approach to s. 1 was applied in Osborne v. Canada (Treasury Board), 1991 CanLII 60 (SCC), [1991] 2 S.C.R. 69, in which this Court struck down a prohibition on partisan political activity by public servants. Much like the legislation at issue in this appeal, the provision in Osborne banned “all partisan-related work by all public servants, without distinction either as to the type of work, or as to their relative role, level or importance in the hierarchy of the public servant” (per Sopinka J., at p. 100 (emphasis added)). Moreover, in Osborne the Court was referred, as in this case, to legislation in other jurisdictions that made distinctions both as to the activity proscribed and the level of public servant, without any weakening of the underlying objective. Based on these factors, Sopinka J. concluded that “[t]he restrictions on freedom of expression in this case are over-inclusive and go beyond what is necessary to achieve the objective of an impartial and loyal civil service” (p. 100). Although the Court differed on whether to strike down the legislation or to read it down in that particular case, its minimum impairment analysis provides unequivocal authority for the current appeal (see, similarly, Rocket v. Royal College of Dental Surgeons of Ontario, 1990 CanLII 121 (SCC), [1990] 2 S.C.R. 232, striking down a restriction on advertising by dentists on grounds that it was overbroad).

62 Turning to the Ontario legislation, my view is that the s. 1 justification suffers, first, from the lack of a recognition of the evolving nature of Ontario agriculture. To the extent the term “family farm” refers to a unique management style characterized by significant family involvement, it may indeed continue to describe the vast majority of farms in Ontario and across Canada. However, to the extent that it treats farm workers as members of that family rather than typical employees, it ignores an increasing trend in Canada towards corporate farming and complex agribusiness. On this point, the Attorney General’s expert himself concedes that “[t]he modern viable family farm no longer consists of 20 acres and a few cows, but typically represents a sophisticated business unit with a minimum capital value of $500,000 to $1,000,000, depending on the commodity and type of operation”. If this is the case, it is not only over-inclusive to perpetuate a pastoral image of the “family farm”, but it may be that certain if not all “family farms” would not be affected negatively by the creation of agricultural associations.

63 The reality is that family involvement does not suffice to alter the essential qualities of an employment relationship; these qualities may include a contract of employment, a consistent wage, regular hours and a hierarchical relationship between employer and employee. Moreover, the traditional family farm is rapidly assuming a less important role in the agricultural sector, as evidenced by increases in non-family farm incorporations, hired farm labour, seasonal workers and average labour costs (see J. White, A Profile of Ontario Farm Labour (March 1997)). Under these circumstances, what the Attorney General for Ontario refers to as an “integration of business and family life” does not, in my view, justify the unqualified and total exclusion of agricultural workers from the LRA. This conclusion echoes that of numerous labour boards legislatures and scholars of labour law, all of whom deny any industrial relations rationale for totally excluding agricultural workers from LRA protection (see, e.g., Wellington Mushroom Farm, supra; South Peace Farms and Oil, Chemical and Atomic Workers International Union, Local No. 9-686, [1977] 1 Can. L.R.B.R. 441; K. Neilson and I. Christie, “The Agricultural Labourer in Canada: A Legal Point of View” (1975), 2 Dal. L.J. 330; Adams, supra, at p. 6-50; Labour Law Casebook Group, Labour and Employment Law: Cases, Materials and Commentary (6th ed. 1998), at pp. 220-21; Beatty, supra, at pp. 91-92).

64 The Attorney General submits that distinguishing various sectors of agriculture requires an impossible line-drawing exercise which the legislature should have the discretion to reject. However, the fact that some legislation includes exceptions for smaller or family-run farms, most notably labour codes in New Brunswick and Quebec, as well as the ALRA itself, suggests that such an exercise is eminently possible, should the legislature choose to undertake it (see Industrial Relations Act, R.S.N.B. 1973, c. I-4, s. 1(5)(a); Labour Code, R.S.Q., c. C-27, s. 21; ALRA). Moreover, it cannot be assumed that a categorical exclusion of agricultural workers removes the need to draw difficult lines. Such an exclusion relies on courts and labour boards to define the meaning of the term “agriculture”, a term that may be as fraught with value judgments as the term “family farm” (see Adams, supra, at pp. 6-49 and 6-50; Neilson and Christie, supra, at pp. 335-41). Thus, while the decision whether to distinguish various sectors of agriculture ultimately rests with the legislature, there is little reason to believe that by totally excluding agricultural workers from the LRA, an impossible line-drawing exercise is avoided.

65 More importantly, the Attorney General offers no justification for excluding agricultural workers from all aspects of unionization, specially those protections that are necessary for the effective formation and maintenance of employee associations. It might be inferred that in order to protect the family farm and ensure the productivity of the farm economy, the legislature felt it necessary to discourage any form of union and to suffer that agricultural workers be exposed to a raft of unfair labour practices. Yet no policy could, in my view, be more repugnant to the principle of least intrusive means. If what is truly sought by s. 3(b) of the LRA is the protection of the family farm, the legislature should at the very least protect agricultural workers from the legal and economic consequences of forming an association. There is nothing in the record to suggest that such protection would pose a threat to the family farm structure, and if demonstrated that it would in some cases, the legislature could create the appropriate exceptions. I am of the view that the wholesale exclusion of agricultural workers from the LRA is not a reasonable limit on freedom of association and that it is not necessary to balance the effects of this exclusion against its stated purpose.

C. Remedy

66 To the extent they substantially impede the effective exercise of the freedom of association, both the LRESLAA and s. 3(b) of the LRA must be declared contrary to the Charter. Given the nature of these enactments, however, determining the appropriate remedy is not without difficulty. First, the respondents point out that the precise effect of striking down the LRESLAA would be to re-enact the statute it repealed, namely, the ALRA. As this Court is not in a position to enact such detailed legislation, nor to confer constitutional status on a particular statutory regime, I prefer to strike down the LRESLAA to the extent that it gives effect to the exclusion clause of the LRA. The precise effect of this remedy is to strike down that exclusion clause, which is the alternate remedy sought by the appellants. This remedy presents its own problems, as it obliges the legislature to extend the full panoply of collective bargaining rights in the LRA to agricultural workers. As such action is not necessarily mandated by the principles of this case, I would suspend the declarations of invalidity for 18 months, allowing amending legislation to be passed if the legislature sees fit to do so. Such a remedy was discussed by Lamer C.J. in Rodriguez v. British Columbia (Attorney General), 1993 CanLII 75 (SCC), [1993] 3 S.C.R. 519, at p. 570, in response to legislation he had found to be over-inclusive:

. . . this Court has recognized that an immediate declaration of invalidity is not always advisable, especially where, as here, the provision pursues an important objective but is over-inclusive: were this Court to strike down the provision effective immediately, those whom the government could protect constitutionally with a more tailored provision, and who indeed should be protected, would be left unprotected. This would clearly pose a “potential danger to the public”. . . .

67 This raises the question of whether s. 2(d) requires that a minimum level of LRA protection be extended to agricultural workers. As implied by Rodriguez, supra, the Charter only obliges the legislature to provide a statutory framework that is consistent with the principles established in this case, including both the s. 2(d) and s. 1 analysis. In my view, these principles require at a minimum a regime that provides agricultural workers with the protection necessary for them to exercise their constitutional freedom to form and maintain associations. The record shows that the ability to establish, join and maintain an agricultural employee association is substantially impeded in the absence of such statutory protection and that this impediment is substantially attributable to the exclusion itself, rather than to private action exclusively. Moreover, the freedom to establish, join and maintain an agricultural employee association lies at the core of s. 2(d) of the Charter; the appellants’ claim is ultimately grounded in this non-statutory freedom. For these reasons, I conclude that at minimum the statutory freedom to organize in s. 5 of the LRA ought to be extended to agricultural workers, along with protections judged essential to its meaningful exercise, such as freedom to assemble, to participate in the lawful activities of the association and to make representations, and the right to be free from interference, coercion and discrimination in the exercise of these freedoms.

68 In choosing the above remedy, I neither require nor forbid the inclusion of agricultural workers in a full collective bargaining regime, whether it be the LRA or a special regime applicable only to agricultural workers such as the ALRA. For example, the question of whether agricultural workers have the right to strike is one better left to the legislature, especially given that this right was withheld in the ALRA (s. 10). Rather than adjudicate such issues at the remedy stage, I adopt the position of Cory and Iacobucci JJ. in Delisle, supra, at para. 151, which is to fashion a remedy according to the nature of the appellant’s claim:

The Court has been asked in this case to rule upon whether the impugned [provision] is unconstitutional because of its anti-associational purpose. We have found that the exclusion of RCMP members from the basic associational protections in the PSSRA does have this purpose and violates the Charter, yet because of the manner in which the appellant has articulated his claim we have done so without being required to decide whether a Charter violation results from the total exclusion of RCMP members from the PSSRA’s collective bargaining regime. As explained by Sopinka J. in PIPSC, supra, at p. 405, it may be that such a total exclusion could interfere with the ability of employees to associate, and thus infringe the Charter’s freedom of association guarantee. We do not believe that it is appropriate to decide, at the remedy stage of the analysis, whether it is constitutionally permissible to exclude RCMP members entirely from a collective bargaining regime. Moreover, we do not wish to prejudge the question of whether Parliament may wish to extend limited collective bargaining rights to RCMP members. [Emphasis added.]

69 Should a claim for inclusion arise in the future, the threshold question will be whether the provision relates to an activity falling within the framework established by the labour trilogy or that otherwise furthers the purpose of s. 2(d) of the Charter. If this threshold is crossed, the question becomes whether excluding agricultural workers from the provision in question substantially impedes this activity either in purpose or effect. If the effect of the exclusion is impugned, the claimant’s position should be assessed in light of the considerations discussed above.

VII. Conclusion

70 For the foregoing reasons, I would allow the appeal with costs throughout. I would declare the LRESLAA unconstitutional to the extent that it gives effect to s. 3(b) of the LRA, and I would declare s. 3(b) of the LRA unconstitutional. I would suspend such declarations for a period of 18 months. I would answer the constitutional questions as follows:

1. Does s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Canadian Charter of Rights and Freedoms; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

Section 80 of the Labour Relations and Employment Statute Law Amendment Act limits freedom of association guaranteed by s. 2(d) of the Charter to the extent that its effect is to re-subject agricultural workers to the exclusion clause found in s. 3(b) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A. For this reason, it is not necessary to answer the s. 15(1) question.

2. Does s. 3(b) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Charter; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

Section 3(b) of the Labour Relations Act, 1995 limits the right of agricultural workers to freedom of association guaranteed by s. 2(d) of the Charter. For this reason, it is not necessary to answer the s. 15(1) question.

3. If the answer to any part of questions 1 or 2 is in the affirmative, is the limitation nevertheless justified under s. 1 of the Charter?

No.

The following are the reasons delivered by

71 L’Heureux-Dubé J. -- I have read the reasons of Bastarache J. I believe that this case can be resolved on simpler grounds. I will therefore outline the reasoning upon which I base my opinion.

72 At the heart of this case is the question of whether the right of agricultural workers in Ontario to associate in order to pursue common goals and their equality rights have been violated by s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1 (“LRESLAA”), and s. 3(b) of the Ontario Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A (“LRA”). The issues in the present appeal are (1) whether the government of Ontario has a positive obligation to protect the appellants’ constitutionally guaranteed rights under s. 2(d) of the Canadian Charter of Rights and Freedoms; and (2) whether the impugned legislation violates s. 15(1) of the Charter.

73 My colleague has set out the facts in this case as well as a description of its judicial history. Except where otherwise expanded upon or noted, I adopt this factual background.

74 In 1994, a specialized legislative regime, the Agricultural Labour Relations Act, 1994, S.O. 1994, c. 6 (“ALRA”), was enacted granting agricultural workers in Ontario protection against unfair labour practices. The United Food and Commercial Workers Union (“UFCW”) was established shortly after the enactment of the ALRA and was certified as the bargaining agent for approximately 200 workers at the Highline Produce Limited mushroom factory farm in Leamington, Ontario. During the period that the ALRA was in effect, the UFCW filed two further certification applications, one for the workers at Kingsville Mushroom Farm Inc., and the other for the workers at the respondent Fleming Chicks.

75 On November 10, 1995, a new government repealed the ALRA and replaced it with legislation that mandated the dissolution of agricultural labour unions. The new LRA excluded persons in agriculture from application of the Act, including statutory protection against unfair labour practices.

76 In Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995, speaking for a majority of the Court, I addressed the issue of positive government obligations in the context of Charter analysis by first noting, at p. 1038 (quoting Reference re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313 (“Alberta Reference”), per Dickson C.J., dissenting, at p. 361):

Section 2 of the Charter protects fundamental “freedoms” as opposed to “rights”. Although these two terms are sometimes used interchangeably, a conceptual distinction between the two is often drawn. “Rights” are said to impose a corresponding duty or obligation on another party to ensure the protection of the right in question whereas “freedoms” are said to involve simply an absence of interference or constraint. This conceptual approach to the nature of “freedoms” may be too narrow since it fails to acknowledge situations where the absence of government intervention may in effect substantially impede the enjoyment of fundamental freedoms (e.g., regulations limiting the monopolization of the press may be required to ensure freedom of expression and freedom of the press). [Emphasis added by L’Heureux-Dubé J. in Haig.]

77 Having set the stage, I then proceeded to develop an argument in Haig that has direct application to this case, stating, at p. 1039:

. . . distinctions between “freedoms” and “rights”, and between positive and negative entitlements, are not always clearly made, nor are they always helpful. One must not depart from the context of the purposive approach articulated by this Court in R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295. Under this approach, 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. This might, for example, take the form of legislative intervention aimed at preventing certain conditions which muzzle expression, or ensuring public access to certain kinds of information.

In the proper context, these may perhaps be relevant considerations leading a court to conclude that positive governmental action is required. [Emphasis added.]

78 In Native Women's Assn. of Canada v. Canada, 1994 CanLII 27 (SCC), [1994] 3 S.C.R. 627, at p. 667, I pointed out that Haig also stands for the proposition that while the government may have been under no constitutional obligation to provide for the right to a referendum under s. 2(b) of the Charter, once the government decides to provide a specific platform of expression, it must do so in a manner consistent with the Charter. I refer specifically to p. 1041 of Haig:

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 [a] ground prohibited under s. 15 of the Charter. [Emphasis added.]

79 Turning to the labour relations context in Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989, a case dealing with s. 2(d) rights of government employees, I noted, at para. 7:

. . . in cases where the employer does not form part of government, there exists no Charter protection against employer interference. In such a case, it might be demonstrated that the selective exclusion of a group of workers from statutory unfair labour practice protections has the purpose or effect of encouraging private employers to interfere with employee associations. It may also be that there is a positive obligation on the part of governments to provide legislative protection against unfair labour practices or some form of official recognition under labour legislation, because of the inherent vulnerability of employees to pressure from management, and the private power of employers, when left unchecked, to interfere with the formation and administration of unions.

80 This case is one where I believe there is a positive obligation on the government to provide legislative protection against unfair labour practices.

81 In 1943, the Ontario Legislature enacted a protective shield for workers, namely, the Collective Bargaining Act, 1943, S.O. 1943, c. 4. But this shield was missing panels, including the protection of agricultural workers. The shield’s pattern had been imported from American legislation. In 1994, 12 years after the enactment of the Charter, the government attempted to make the shield whole by enacting the ALRA. The patched shield was intact for 17 months. In 1995, the legislature removed the panel and reinstated, in a new labour statute, a modified version of the original exclusion clause. Thus, under the often harsh labour relations climate, agricultural workers were once again made to stand out as a result of their exclusion from the protection offered by the statutory shield, in stark contrast to the vast majority of workers who were protected by the LRA.

82 I start with an examination of the legislative history and factual background, an analysis intended to complement the narrative included in my colleague’s reasons. It should be noted that both the decision of the Ontario Court (General Division) ((1997), 1997 CanLII 12345 (ON SC), 155 D.L.R. (4th) 193, per Sharpe J. (as he then was)) and that of the Ontario Court of Appeal ((1999), 1999 CanLII 18653 (ON CA), 182 D.L.R. (4th) 471), predate this Court’s decision in Delisle, supra.

I. Legislative History and Factual Background

83 Ontario was the first jurisdiction in Canada “to adopt a fully-fledged collective bargaining statute” (H. W. Arthurs et al., Labour Law and Industrial Relations in Canada (4th ed. 1993), at para. 90). On April 14, 1943, the Collective Bargaining Act, 1943 was assented to in Ontario. At s. 24, it listed the categories of employees excluded from the Act, including “domestic servants”, “members of any police force”, certain other public employees, and “the industry of farming”.

84 The Collective Bargaining Act, 1943 was modelled on a statute enacted by the United States Congress in 1935, the National Labor Relations Act, July 5, 1935, c. 372, 49 Stat. 449 (29 U.S.C. §§ 151 to 169) (called the “Wagner Act” after its sponsor, Senator Wagner of New York). Senator Wagner considered this ground-breaking statute more “than a weapon against the disruption of industry by labor-management disputes”. He envisaged it “as an ‘affirmative vehicle’ for economic and social progress” (C. J. Morris, ed., The Developing Labor Law (1971), at p. 27).

85 Senator Wagner’s address to Congress (79 Cong. Rec. 7565 (1935)), was prescient in its use of terminology familiar to current Canadian Charter jurisprudence: “Caught in the labyrinth of modern industrialism and dwarfed by the size of corporate enterprise, [the employee] can attain freedom and dignity only by cooperation with others of his group” (quoted in Morris, supra, at p. 27 (emphasis added)).

86 The Wagner Act excluded certain categories of employees, including agricultural workers. Several of these exclusions are reflected in the Collective Bargaining Act, 1943.

87 On April 6, 1944, the Collective Bargaining Act, 1943 was repealed and replaced with The Labour Relations Board Act, 1944, S.O. 1944, c. 29 (“LRBA”). At s. 10(a), the LRBA affirmed that it did not apply to “the industry of farming”.

88 The LRBA was repealed and replaced with The Labour Relations Act, 1948, S.O. 1948, c. 51, which was subsequently repealed and replaced with The Labour Relations Act, 1950, S.O. 1950, c. 34. At s. 2 of the 1950 Act, the restriction was amended to read:

2. This Act does not apply,

(b) to any person employed in agriculture, horticulture, hunting or trapping;

89 On April 12, 1960, clause b of s. 2 was amended (S.O. 1960, c. 54, s. 1) by striking out “horticulture” in the first line and adding a new clause bb:

2. This Act does not apply,

(b) to any person employed in agriculture, hunting or trapping;

(bb) to any person, other than an employee of a municipality or a person employed in silvaculture [sic], who is employed in horticulture by an employer whose primary business is agriculture or horticulture.

90 The issues surrounding the statutory provision at the heart of this case were referred to indirectly in Cuddy Chicks Ltd. v. Ontario (Labour Relations Board), 1991 CanLII 57 (SCC), [1991] 2 S.C.R. 5, where this Court examined the jurisdiction of the Ontario Labour Relations Board (“OLRB”) to determine the constitutionality of s. 2(b) of the Labour Relations Act, R.S.O. 1980, c. 228 (now s. 3(b) of the LRA) in the course of proceedings before the OLRB. One of the issues on appeal was whether the Ontario Court of Appeal erred in holding that the OLRB had jurisdiction to decide the constitutional validity of s. 2(b) of its enabling statute by applying the Charter as part of its duty to consider statutes bearing on proceedings before it.

91 The underlying case in Cuddy Chicks can be summarized as follows. In April 1987, the United Food and Commercial Workers International Union, Local 175, filed an application for certification before the OLRB relating to employees at the chicken hatchery of Cuddy Chicks Limited. Section 2(b) of the 1980 Labour Relations Act stated that the Act does not apply “to a person employed in agriculture”. The OLRB noted at para. 44 of its reasons in Cuddy Chicks Ltd., [1988] O.L.R.B. Rep. May 468, that it was the union, not the employees, that made the Charter application. On filing the application, the union gave notice that, if the employees were found to be agricultural employees, it would request the OLRB to hold s. 2(b) invalid as being contrary to s. 2(d) and s. 15(1) of the Charter. In affirming that the employees were indeed so employed, the OLRB stated, at para. 9, that:

There is no doubt that the hatchery is a highly mechanized, technologically sophisticated operation and that the employees in many respects work in factory-like conditions with set shifts, year-round employment and the benefits and disciplinary provisions similar to or the same as one would expect to find in a factory. We accept respondent counsel’s submission that agriculture has become highly technological and commercial, but that that does not make those activities non-agricultural: Wellington Mushroom Farm, [1980] OLRB Rep. May 813. It is thus the nature of the activities and not the way they are performed or the tools by which they are performed that is relevant. [Emphasis added.]

92 On the issue of the OLRB’s jurisdiction to hear the Charter challenge, a majority of the Board decided that it had jurisdiction and directed the Registrar to set dates for the panel to hear evidence and argument on the union’s Charter challenge to s. 2(b) of the 1980 Act. This hearing was postponed due to litigation launched by Cuddy Chicks Limited, which eventually came before this Court.

93 In Cuddy Chicks, supra, this Court ruled that the OLRB had jurisdiction to examine whether s. 2(b) of the 1980 Labour Relations Act was contrary to the Charter.

94 The matter was not carried forward, however, because of the imminence of legislative change. In Cuddy Chicks Ltd., [1992] O.L.R.D. No. 1170 (QL), the Board reported, at para. 1, that:

The applicant and respondent, by letter dated March 23, 1992, have jointly requested the Board to hold this matter in abeyance, in light of the amendments currently being considered for the Act, in particular as it affects the exclusion of “agricultural” employees in section 2(b).

95 In June 1992, the Report of the Task Force on Agricultural Labour Relations: Report to the Minister of Labour was released. The recommendations contained in this report influenced legislation subsequently introduced by the Ontario government, the ALRA, which was assented to on June 23, 1994. In recognition of certain specific concerns about the impact of the extension of statutory labour rights to the agricultural sector, the ALRA prohibited strikes and lockouts, substituting in their place a dispute resolution process, the final stage of which was binding final offer selection by an arbitration board.

96 The ALRA was repealed by a newly elected government in 1995. The LRESLAA, which was assented to on November 10, 1995, replaced the predecessor Labour Relations Act and related amendments, and repealed the ALRA.

97 The LRESLAA also stipulated that any agreements certified under the ALRA were henceforth terminated, as were any certification rights of trade unions. The LRESLAA explicitly prohibited employers from reprisals against workers on account of union activity under the ALRA. The effect of the LRESLAA was to subject agricultural workers to the exclusion clause of the LRA (s. 3(b)). Alberta is the only other Canadian province with labour relations legislation incorporating an unqualified exclusion of agricultural workers.

98 The LRA is a substantial statute implementing a comprehensive labour relations regime. Its comprehensiveness can perhaps be appreciated by contrasting its 169 sections with the 27 sections of the Collective Bargaining Act, 1943.

99 The LRA features an expansive definition of “agriculture” at s. 1(1):

“agriculture” includes farming in all its branches, including dairying, beekeeping, aquaculture, the raising of livestock including non‑traditional livestock, furbearing animals and poultry, the production, cultivation, growing and harvesting of agricultural commodities, including eggs, maple products, mushrooms and tobacco, and any practices performed as an integral part of an agricultural operation, but does not include anything that was not or would not have been determined to be agriculture under section 2 of the predecessor to this Act as it read on June 22, 1994;

100 The exclusionary section of the LRA largely replicates the language of its predecessor:

3. This Act does not apply,

(a) to a domestic employed in a private home;

(b) to a person employed in agriculture, hunting or trapping;

(c) to a person, other than an employee of a municipality or a person employed in silviculture, who is employed in horticulture by an employer whose primary business is agriculture or horticulture;

(d) to a member of a police force within the meaning of the Police Services Act;

(e) except as provided in Part IX of the Fire Protection and Prevention Act, 1997, to a person who is a firefighter within the meaning of subsection 41(1) of that Act;

(f) to a member of a teachers’ bargaining unit established by Part X.1 of the Education Act, except as provided by that Part, or to a supervisory officer, a principal or a vice‑principal;

(g) to a member of the Ontario Provincial Police Force;

(h) to an employee within the meaning of the Colleges Collective Bargaining Act;

(i) to a provincial judge; or

(j) to a person employed as a labour mediator or labour conciliator. [Emphasis added.]

101 It is worth noting that apart from the first three excluded categories, the remaining targets of the exclusion belong to groups that enjoy legal and social recognition and respect. Members of these groups must undergo rigorous selection and training processes before gaining admittance, and they generally enjoy stable employment conditions, comfortable salaries and benefits during their working lives, and adequate pension plans upon retirement. In many cases they also enjoy the protection and services of an established and dedicated union, which can engage in collective bargaining pursuant to other statutes. A parallel can be drawn between the situation of members of these relatively privileged and correspondingly less vulnerable groups, and the RCMP officers in Delisle, supra.

102 In stark contrast, entry into the first three excluded categories generally requires little if any formal training. And, at least insofar as agricultural workers are concerned, working conditions are characterized by long hours, low wages, little job security or social recognition, and few employment benefits beyond those strictly mandated by law (see the affidavit of Professor Judy Fudge from Osgoode Hall Law School). As Sharpe J. noted in his reasons, “agricultural workers are a disadvantaged group. They are poorly paid, face difficult working conditions, have low levels of skill and education, low status and limited employment mobility” (p. 216).

103 I would like to make explicit reference to the fact that in these reasons we are not deciding on the rights, or lack thereof, of foreign seasonal agricultural workers and their families, who are regulated under federal legislation.

II. Charter Analysis

104 At issue in the present appeal is whether the impugned legislation violates the appellants’ freedom of association and equality rights guaranteed under the Charter.

105 In New Brunswick Broadcasting Co. v. Nova Scotia (Speaker of the House of Assembly), 1993 CanLII 153 (SCC), [1993] 1 S.C.R. 319, McLachlin J. (as she then was), stated that “[t]he Charter has changed the balance of power between the legislative branch and the executive on the one hand, and the courts on the other hand, by requiring that all laws and government action must conform to the fundamental principles laid down in the Charter” (p. 389 (emphasis added)).

106 As stated by Dickson J. (as he then was), for this Court in Hunter v. Southam Inc., 1984 CanLII 33 (SCC), [1984] 2 S.C.R. 145, at p. 156, the intent of the Charter is to constrain government action:

I begin with the obvious. The Canadian Charter of Rights and Freedoms is a purposive document. Its purpose is to guarantee and to protect, within the limits of reason, the enjoyment of the rights and freedoms it enshrines. It is intended to constrain governmental action inconsistent with those rights and freedoms. [Emphasis added.]

107 This Court has, on several occasions, set forth the guidelines to be employed in construing Charter provisions. In R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103, at p. 119, Dickson C.J. stated that “[t]o identify the underlying purpose of the Charter right in question . . . it is important to begin by understanding the cardinal values it embodies”. In R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295, at p. 344, Dickson J. stated:

In Hunter v. Southam Inc., 1984 CanLII 33 (SCC), [1984] 2 S.C.R. 145, this Court expressed the view that the proper approach to the definition of the rights and freedoms guaranteed by the Charter was a purposive one. The meaning of a right or freedom guaranteed by the Charter was to be ascertained by an analysis of the purpose of such a guarantee; it was to be understood, in other words, in the light of the interests it was meant to protect.

In my view this analysis is to be undertaken, and the purpose of the right or freedom in question is to be sought by reference to the character and the larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concepts enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter. The interpretation should be, as the judgment in Southam emphasizes, a generous rather than a legalistic one, aimed at fulfilling the purpose of the guarantee and securing for individuals the full benefit of the Charter’s protection. At the same time it is important not to overshoot the actual purpose of the right or freedom in question, but to recall that the Charter was not enacted in a vacuum, and must therefore, as this Court’s decision in Law Society of Upper Canada v. Skapinker, 1984 CanLII 3 (SCC), [1984] 1 S.C.R. 357, illustrates, be placed in its proper linguistic, philosophic and historical contexts. [Emphasis added.]

108 As stated by Laycraft J.A. of the Court of Appeal of Alberta in R. v. Big M Drug Mart Ltd. (1983), 1983 ABCA 268, 49 A.R. 194, “the Charter has both a positive and a negative impact. Acting positively it grants and guarantees rights to Canadians; negatively it imposes a corresponding and opposite limit on the power of government” (p. 203).

109 In Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493, at para. 60, the majority discussed the scope of government obligations under s. 32 of the Charter in the context of underinclusive legislation:

The relevant subsection, s. 32(1)(b), states that the Charter applies to “the legislature and government of each province in respect of all matters within the authority of the legislature of each province”. There is nothing in that wording to suggest that a positive act encroaching on rights is required; rather the subsection speaks only of matters within the authority of the legislature. Dianne Pothier has correctly observed that s. 32 is “worded broadly enough to cover positive obligations on a legislature such that the Charter will be engaged even if the legislature refuses to exercise its authority” (“The Sounds of Silence: Charter Application when the Legislature Declines to Speak” (1996), 7 Constitutional Forum 113, at p. 115). The application of the Charter is not restricted to situations where the government actively encroaches on rights. [Emphasis in original.]

110 The original Acts were drafted and amended in the pre-Charter era. To develop an ex post facto finding of Charter-complying legislative intent would be a speculative exercise. The same cannot be said about the drafting of the 1995 legislation, which must meet the constitutional guarantees set out in the Charter. The entrenchment of the Charter marks an important date in our legal evolution.

A. Section 2(d) Analysis

111 The appellants claim that the LRESLAA and the LRA violate agricultural workers’ freedom of association under s. 2(d) of the Charter. The freedom of association is a fundamental freedom under the Charter. In the Alberta Reference, supra, Dickson C.J. elaborated on the role of freedom of association as follows, at p. 334:

Freedom of association is the freedom to combine together for the pursuit of common purposes or the advancement of common causes. It is one of the fundamental freedoms guaranteed by the Charter, a sine qua non of any free and democratic society, protecting individuals from the vulnerability of isolation and ensuring the potential of effective participation in society. In every area of human endeavour and throughout history individuals have formed associations for the pursuit of common interests and aspirations. Through association individuals are able to ensure that they have a voice in shaping the circumstances integral to their needs, rights and freedoms. [Emphasis added.]

112 In the same case, McIntyre J. stated at p. 408 that the “fundamental purpose of freedom of association . . . [is] to permit the collective pursuit of common goals”.

113 In Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211, at p. 252, Wilson J. reviewed the analysis on freedom of association in the Alberta Reference, supra, and concluded that:

[I]n construing the purpose behind s. 2(d) this Court was unanimous in finding that freedom of association is meant to protect the collective pursuit of common goals. This reading of the purpose behind the guarantee of freedom of association has been confirmed in more recent cases. For instance, s. 2(d) was considered again in the labour relations context in Professional Institute of the Public Service of Canada v. Northwest Territories (Commissioner), 1990 CanLII 72 (SCC), [1990] 2 S.C.R. 367 (“P.I.P.S.”). [Emphasis added.]

114 Thus, I agree with my colleague Bastarache J. that the purpose of s. 2(d) is to protect the collective pursuit of common goals. With respect, however, I do not agree with his assertion that the right not to associate is protected under s. 2(d) of the Charter (see R. v. Advance Cutting & Coring Ltd., [2001] 3 S.C.R. 209, 2001 SCC 70), but this right is not implicated in the present case, and I say no more about it.

115 Worker organizations are a powerful and vibrant example of the collective pursuit of common goals. In the Alberta Reference, supra, at pp. 334-35, Dickson C.J. commented on the fundamental importance of the freedom to associate in the context of labour relations:

Freedom of association is the cornerstone of modern labour relations. Historically, workers have combined to overcome the inherent inequalities of bargaining power in the employment relationship and to protect themselves from unfair, unsafe, or exploitative working conditions. As the United States Supreme Court stated in N.L.R.B. v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937), at p. 33:

Long ago we stated the reason for labour organizations. We said that they were organized out of the necessities of the situation; that a single employee was helpless in dealing with an employer; that he was dependent ordinarily on his daily wage for the maintenance of himself and family; that if the employer refused to pay him the wages that he thought fair, he was nevertheless unable to leave the employ and resist arbitrary and unfair treatment; . . .

The “necessities of the situation” go beyond, of course, the fairness of wages and remunerative concerns, and extend to matters such as health and safety in the work place, hours of work, sexual equality, and other aspects of work fundamental to the dignity and personal liberty of employees. [Emphasis added.]

116 The above comments reflect universal aspirations, and it is not surprising that Senator Wagner’s words, supra, at para. 85, closely resemble the language employed by Dickson C.J. close to half a century later.

117 In Delisle, Cory and Iacobucci JJ., writing in dissent, stated, at para. 67:

The Court has also acknowledged the inherent vulnerability and inequality of the individual employee in the workplace in the face of management. In Slaight Communications Inc. v. Davidson, 1989 CanLII 92 (SCC), [1989] 1 S.C.R. 1038, at p. 1051, Dickson C.J. concluded, on behalf of the majority of the Court, that employees are a vulnerable group in Canadian society. In Wallace, supra, Iacobucci J. noted that this vulnerability is underscored by the very importance which our society attaches to employment. He emphasized the inequality of bargaining power and information between employees and employers, noting that this power imbalance is not limited to the context of the employment contract proper, but rather affects “virtually all facets of the employment relationship”: para. 92.

118 In the Alberta Reference, supra, Dickson C.J., while commenting about the freedom of association, stated, at p. 368:

Work is one of the most fundamental aspects in a person’s life, providing the individual with a means of financial support and, as importantly, a contributory role in society. A person’s employment is an essential component of his or her sense of identity, self‑worth and emotional well‑being. Accordingly, the conditions in which a person works are highly significant in shaping the whole compendium of psychological, emotional and physical elements of a person’s dignity and self respect.

119 In Lavigne, supra, at p. 241, Wilson J. reiterated that the Charter is concerned with the purpose of state action with regards to Charter rights as well as their effect on an individual’s guaranteed rights or freedoms. The oft-quoted passage from Dickson J. in Big M Drug Mart, supra, summarizes the point, at p. 331:

In my view, both purpose and effect are relevant in determining constitutionality; either an unconstitutional purpose or an unconstitutional effect can invalidate legislation. All legislation is animated by an object the legislature intends to achieve. This object is realized through the impact produced by the operation and application of the legislation. Purpose and effect respectively, in the sense of the legislation’s object and its ultimate impact, are clearly linked, if not indivisible. Intended and actual effects have often been looked to for guidance in assessing the legislation’s object and thus, its validity. [Emphasis added.]

Purpose of the Exclusion Clause

120 The purpose of s. 3(b) of the LRA is clear: to prevent agricultural workers from unionizing. This purpose infringes s. 2(d) of the Charter.

121 In Delisle, supra, at para. 6, after expressing my agreement with Cory and Iacobucci JJ. with regard to their views on the importance of freedom of association and the inherent vulnerability of workers in the face of management, I added that:

The unique context of labour relations must always be considered in constitutional claims in this area, and the right to freedom of association must take into account the nature and importance of labour associations as institutions that work for the betterment of working conditions and the protection of the dignity and collective interests of workers in a fundamental aspect of their lives: employment. The contextual approach to Charter analysis must also take into account the history of the need for government intervention to make effective the rights of workers to associate together. I agree with my colleagues that both intrinsic and extrinsic sources are admissible and significant in determining legislative purpose and effects, and with their comments on the fact that an invalid purpose is sufficient to find a violation of a Charter right. [Emphasis in original.]

122 In Delisle I based my conclusion that no s. 2(d) breach had taken place in part on the fact that the evidence “[did] not show that the object of the exclusion was to impede the formation of independent employee associations”, but rather that “the exclusion stemmed from a desire not to grant RCMP members all of the rights contemplated by the legislation and access to the particular remedies contained within it” (para. 5). In that case, the employer was the government and thus it was possible to presume that the government/employer knew of its responsibility under the Charter to respect the workers’ associational rights. As a result, it was possible to assume that the purpose of the impugned law was to deny RCMP employees certain statutory benefits, but still respect their basic right to associate. That is not the case here, where the employers are not part of government, and therefore their anti-associational acts could not be challenged under the Charter. Accordingly, there can be no presumption that the Ontario government expected that the Charter would protect farm workers’ basic freedom to associate.

123 I respectfully disagree with my colleague when he argues that there was no clear evidence of intent on the part of the government of Ontario to breach the s. 2(d) rights of the appellants. It is difficult to countenance such a thesis in light of the factual record which includes not only the direction to terminate existing associations contained in the LRESLAA and the s. 3(b) exclusion contained in the LRA, but also the repeated instances where government officials made it clear that the new Act’s intent was to hinder union-related activities in the agricultural sector, including comments made to members of the provincial legislature before they voted on the LRESLAA.

124 To interpret the factual record so narrowly by arguing that it was not clear whether these statements were aimed solely at curtailing unprotected rights to engage in collective bargaining or whether they were also aimed at curtailing Charter-protected associational rights defies purposive Charter jurisprudence. It appears to me that in situations, such as this one, where there are strong indicia that the intent of a legislative initiative is to curtail a Charter right, the onus shifts to the government to justify the breach under s. 1.

125 I believe that, in the circumstances of this case, the proper approach is the one taken by Dickson J. in Big M Drug Mart, supra, at p. 334:

. . . I agree with the respondent that the legislation’s purpose is the initial test of constitutional validity and its effects are to be considered when the law under review has passed or, at least, has purportedly passed the purpose test. If the legislation fails the purpose test, there is no need to consider further its effects, since it has already been demonstrated to be invalid. Thus, if a law with a valid purpose interferes by its impact, with rights or freedoms, a litigant could still argue the effects of the legislation as a means to defeat its applicability and possibly its validity. In short, the effects test will only be necessary to defeat legislation with a valid purpose; effects can never be relied upon to save legislation with an invalid purpose. [Emphasis added.]

126 Charter litigation decisions cannot be made in a factual vacuum. However, it is important to assess carefully on whom the burden of proof should lie, and the degree of proof required. Cory and Iacobucci JJ. in Delisle present a helpful outline, at para. 76:

A Charter claimant who seeks to establish that impugned legislation infringes a Charter right or freedom by virtue of its purpose bears the onus of establishing the alleged invalid purpose on a balance of probabilities. The ordinary rules of evidence applicable in civil trials apply. Accordingly, it cannot be assumed that the purpose of a law is invalid solely because an invalid purpose is a plausible purpose of the law. There must be clear evidence that an invalid purpose is probable. In addition, the evidence must rebut the presumption of constitutionality. That is, if there are two equally probable purposes for the impugned legislation, and one of these purposes is valid and is not inextricably linked to the invalid purpose, then the valid purpose is presumed to apply: Slaight Communications, supra, at p. 1078, per Lamer J. (as he then was); Canada (Attorney General) v. Mossop, 1993 CanLII 164 (SCC), [1993] 1 S.C.R. 554, at pp. 581‑82, per Lamer C.J. However, where the Charter claimant is able to adduce a preponderance of evidence of the invalid purpose, the presumption of constitutionality is rebutted and the court is required to find an infringement of the Charter. [Emphasis added.]

127 In assessing whether the purpose of a legislative provision is constitutional, “the court [should] consider only the purpose of the provision itself and not the broader purpose of the surrounding legislation as a whole” (Delisle, supra, per Cory and Iacobucci JJ., at para. 78). Thus, in the present appeal, the essential issue under s. 2(d) of the Charter is whether the purpose of s. 80 of the LRESLAA and s. 3(b) of the LRA infringes freedom of association, not whether the legislation as a whole does so. Both intrinsic and extrinsic sources are admissible and important in determining legislative purpose.

128 The evidence in this case leads me to conclude that this is one of those rare cases in which the Ontario Legislature’s purpose in enacting a legislative provision must be found to infringe the Charter. There is clear evidence that on a balance of probabilities the Legislature’s purpose in enacting the impugned s. 3(b) was to ensure that persons employed in agriculture “remained vulnerable to management interference with their associational activities, in order to prevent the undesirable consequences which it was feared would result from [agricultural workers’] labour associations” (Delisle, supra, per Cory and Iacobucci JJ., at para. 80).

129 There were several official announcements as to the purpose of s. 3(b). These can be categorized as announcements where the concept of protecting the “family farm” from unionization featured prominently, and those where incompatibility between agriculture and unionization was cited as the reason for the repeal of the ALRA.

130 The following fall into the first category: (i) a statement by the Ontario Minister of Labour in the Legislature on October 4, 1995, upon introducing the LRESLAA, that “[t]his action . . . recognizes that unionization of the family farm has no place in Ontario’s key agricultural sector”; and (ii) a statement by the Ontario Minister of Agriculture, Food and Rural Affairs also in the Legislature and on the same date that “[o]ur farmers, who are on the agrifood industry’s front lines, are looking to us to help them maintain their competitive edge in the new global marketplace. . . . [T]he Agricultural Labour Relations Act is aimed directly at unionizing the family farm. We do not believe in the unionization of the family farm” (Legislative Assembly of Ontario, Official Report of Debates, October 4, 1995, at pp. 99-100).

131 In the second category, the statements refer to agriculture in general, without specific reference to family farms. First, a media kit released by the government gave the following explanation as to why the ALRA was being repealed: “The horticulture and agriculture sectors are extremely sensitive to time and to climate conditions as these directly affect production of many agricultural commodities. For this reason, these sectors would have great difficulty adapting to the presence of unions”. Second, on January 17, 1996, after the enactment of the LRESLAA, the Minister of Labour responded in a letter that “[t]he Government repealed Bill 91 because of the Government’s view that unionization in the agricultural sector is incompatible with the unique characteristics of that sector”.

132 Contrasting the first statement from the Ontario Minister of Labour from the latter one, it seems that if the purpose of the LRESLAA was to protect the family farm from unionization, then passage of a statute prohibiting all unionization in Ontario’s agricultural sector reflects overreach.

133 In addition, the comments Cory and Iacobucci JJ. make in Delisle, supra, at para. 87, are germane:

In this context, leaving aside altogether the collective bargaining rights and the grievance procedure set out in the PSSRA, the fact that RCMP members are excluded from the application of even these limited associational protections is significant. The PSSRA is modelled upon The Industrial Relations and Disputes Investigation Act. It was enacted at a time when legislative awareness of the fundamental importance of the freedom of employees to associate was high, as evidenced by domestic and international legislation at the time. It is unquestionable that Parliament was aware of the importance of freedom of association for all employees, and of the possibility of protecting this freedom without providing all employees with collective bargaining rights. The symbolism inherent in declining to guarantee to RCMP members even the basic freedom to associate must have been recognized. [Emphasis added.]

134 Substituting agricultural workers for the RCMP and the Ontario Legislature for the Parliament, and applying the facts to our situation leads to a similar conclusion as that reached by Cory and Iacobucci JJ. in Delisle, where, at para. 89, they state:

The key consideration, in examining Parliament’s purpose in excluding members of the RCMP from the PSSRA, is the reason for the decision to exclude. If Parliament’s purpose in excluding a particular employee group from a labour statute was to ensure that the employee group remained vulnerable to management interference with labour association, this is impermissible in light of s. 2(d). Even though the effect of the exclusion may be simply to maintain the status quo of employees whereby they are burdened with the inherent imbalance of power in the employment context, the central consideration is whether Parliament's deliberate decision to exclude flowed from a purpose that is in conflict with the fundamental freedom of employees to associate. It is of some relevance that the status quo in the labour relations context is one of inherent employee vulnerability to management interference with labour associations. It is simply not open to Parliament to enact a statutory provision where the motivation for enacting the provision is anti‑associational, subject of course to s. 1 of the Charter. [Emphasis in original.]

135 As pointed out by Cory and Iacobucci JJ. in Delisle, supra, at para. 102, another factor which may be of assistance would be the existence of any positive effect that the exclusion of persons employed in agriculture may have had on the associational freedom of such persons:

Dickson J. in Big M Drug Mart, supra, emphasized that the effects of impugned legislation need not be looked to if the purpose of the legislation is invalid, and further that even if the effects are looked to and found to be “inoffensive” this fact will not affect a finding that the purpose of the legislation is invalid. Nonetheless, courts may, where appropriate, look to the effects of legislation for assistance in inferring the legislation's purpose, as Dickson J. noted, at p. 331. In particular, where the effects of the impugned legislation are contrary to the invalid purpose alleged by the Charter claimant, a court should weigh the evidence carefully before concluding that the purpose is indeed invalid. In light of the presumption of constitutionality, it is fitting for a court to look for the existence of any such beneficial effects before ruling that the purpose of a law is contrary to the Charter. [Emphasis added; emphasis in original deleted.]

136 The evidence before us fails to reveal any positive effects upon the associational freedom of persons employed in agriculture stemming from their exclusion from the LRA. In fact, I can point to the dearth of employee associations established by persons employed in agriculture in Ontario (see Delisle, at para. 106). The associational record is dismal. But so is the situation of agricultural workers across Canada when it comes to employee associations. Data for 1989 evidences that while, on average, 34.1 percent of Canadian workers belonged to unions, a mere 1.9 percent of workers in agriculture were unionized (Arthurs, supra, at para. 93).

137 The government argues that (a) the banning of unions does not prevent the creation of associations by agricultural workers; and (b) the government did not engage in anti-associational activity, but rather it was private parties who did so.

138 In the context of Ontario’s labour market, it would be disingenuous of the government to argue that it believed that following the enactment of the LRA the freedom of association of agricultural workers would be restricted solely in terms of union activities. The reality of the labour market, which has led to the development of protective labour legislation, indicates that when the protection is removed without any restrictions or qualifications, associational rights are often infringed, or have the potential to be infringed, to an extent not confined to unionization activities. In my view, it cannot be said that the government was unaware, in advance, of this very effect of its legislation.

139 My colleague has concluded that the effect of the enactment of the LRA breaches s. 2(d) of the Charter. While I agree, I also believe that in a situation such as the present one, intent can be imputed on the government. Such cases will be rare and subject to specific circumstances. In this case, (a) the context is that of specialized legislation which seeks to maintain a delicate balance between employees and employers; (b) there was an absolute exclusion from protection; (c) it was in the reasonable contemplation of the government at the time of the enactment that the effect of the exclusion clause would be to affect associational freedoms beyond the realm of unionization, thus breaching s. 2(d) rights.

140 In the present case, the appellants claim that the government has breached the s. 2(d) rights of agricultural workers in Ontario because it has enacted a new labour statute, which leaves them perilously vulnerable to unfair labour practices. The appellants’ claims have merit, particularly when viewed in the context of (a) the recent history of Ontario labour legislation, where the ALRA was enacted only to be repealed by new legislation mandating the dismantling of unions organized under the ALRA and excluding agricultural workers from the LRA without any provisos for protection from unfair labour practices, except for a prohibition against reprisals for agricultural workers who organized under the ALRA; (b) the inherent vulnerability of workers when confronting management; (c) the specific vulnerability and powerlessness of agricultural workers; and (d) the long experience of labour strife which has led to the enactment of statutes protecting workers against unfair labour practices.

141 The difficulties facing employees attempting to organize without the benefit of statutory protection against unfair labour practices would appear to be much greater outside large urban centres, where at least it would be plausible to attend meetings and events without fear of the employer becoming aware of such initiatives. In the countryside, where many workers live on or near their place of employment, where the only meeting hall may be the local Legion or the dance hall in the nearest town, the odds would be stacked against escaping scrutiny by the employers.

142 My colleague makes the point that any interference with s. 2(d) rights must be substantial. While I agree with him that trivial breaches of the Charter should not be given much credence, I believe that we must seek to examine the severity of the Charter breach from the point of view of the party whose rights are affected. As I stated in Egan v. Canada, 1995 CanLII 98 (SCC), [1995] 2 S.C.R. 513, at para. 58, “groups that are more socially vulnerable will experience the adverse effects of a legislative distinction more vividly than if the same distinction were directed at a group which is not similarly socially vulnerable”.

143 In this case, the government argues that by proclaiming that the intent of the legislation was to exclude unionization from the agricultural sector, it merely meant to restrain “collective bargaining” activities. There is no discussion regarding the impact of the LRA on the associational activities in which workers often participate which are separate from the collective bargaining process.

144 This leads us to the issue raised by Dickson C.J. in the Alberta Reference, supra, at pp. 362-63, where he stated, in dissent, that “[i]f freedom of association only protects the joining together of persons for common purposes, but not the pursuit of the very activities for which the association was formed, then the freedom is indeed legalistic, ungenerous, indeed vapid”.

145 We could perhaps draw a useful analogy from the argument made with regards to language rights by my colleague in R. v. Beaulac, 1999 CanLII 684 (SCC), [1999] 1 S.C.R. 768, at para. 20, where he stated that:

Language rights are not negative rights, or passive rights; they can only be enjoyed if the means are provided. This is consistent with the notion favoured in the area of international law that the freedom to choose is meaningless in the absence of a duty of the State to take positive steps to implement language guarantees; see J. E. Oestreich, “Liberal Theory and Minority Group Rights” (1999), 21 Hum. Rts. Q. 108, at p. 112; P. Jones, “Human Rights, Group Rights, and Peoples’ Rights” (1999), 21 Hum. Rts. Q. 80, at p. 83: “[A] right . . . is conceptually tied to a duty”; and R. Cholewinski, “State Duty Towards Ethnic Minorities: Positive or Negative?” (1988), 10 Hum. Rts. Q. 344. [Emphasis added.]

146 Similarly, in the case of agricultural workers in Ontario, the freedom to associate becomes meaningless in the absence of a duty of the state to take positive steps to ensure that this right is not a hollow one.

147 In Committee for the Commonwealth of Canada v. Canada, 1991 CanLII 119 (SCC), [1991] 1 S.C.R. 139, I pointed out, at p. 214:

Rights and freedoms must be nurtured, not inhibited. Vague laws intruding on fundamental freedoms create paths of uncertainty onto which citizens fear to tread, fearing legal sanction. Vagueness serves only to cause confusion and most people will shy from exercising their freedoms rather than facing potential punishment.

148 In the case at bar, citizens employed in agriculture fear not legal sanctions but sanctions from their employers. The absolute removal of LRA protection from agricultural workers created a situation where employees have reason to fear retaliation against associational activity by employers. The Ontario legislation could have qualified the exclusion clause by enjoining retaliatory activity by employers for non-union related associational activity. In light of the reality of the labour market, the failure of the Ontario Legislature to spell out a regime defining which associational activities are to be protected from management retaliation creates a chilling effect for agricultural workers. The concept of chilling effect is premised on the idea that individuals anticipating penalties may hesitate before exercising constitutional rights. In a constitutional democracy, not only must fundamental freedoms be protected from state action, they must also be given “breathing space”.

149 The rights protected under s. 2(d) of the Charter are not confined to the work environment, but cover the full range of activities undertaken by individuals in a vibrant democracy. An employer, however, whose principal interaction with his employees may be confined to the employment situation, may view associational activities as manifestations or precursors of unionization. By taking preventive action, by firing, disciplining or warning targeted employees, with the resulting chilling effect on present and future activities by other employees, an employer would thus be infringing a right considered fundamental in our society. The chilling effect, of course, would not be confined to the employees of that particular employer.

150 The democratic dimensions of the freedom to associate have been rightly referred to in our jurisprudence. Other dimensions are also important. For example, those working in the production of food may, through associational activity, share and enhance their skills and knowledge, an important consideration when one takes into account the fact that agricultural workers are often exposed to dangers from machinery, chemicals and pesticides. Exchanging views on recent developments may ensure that unsafe practices will be identified at an earlier stage. Society would also prize their role as potential stewards over the safety of the products generated and of the environment in general. A statutory regime that may have a chilling effect on such activities would appear to run contrary to the common good.

151 In the context of the exclusion under s. 3(b) of the LRA, the situation is exacerbated because agricultural workers in Ontario, as pointed out by my colleague, are also excluded from employment standards, occupational health and safety, and other protective legislation such as the Tenant Protection Act, 1997, S.O. 1997, c. 24. Section 3(b) of the Tenant Protection Act states that it does not apply to living accommodations whose occupancy is conditional upon the occupant continuing to be employed on a farm, whether or not the accommodations are located on that farm.

152 The chilling effect of the impugned provision has forced agricultural workers to abandon associational efforts and to restrain themselves from further associational initiatives. It may be of assistance to examine the resulting impact on the freedom of association in light of the comments made by Dickson J. in Big M Drug Mart, supra, at pp. 336-37, with regard to freedom of religion:

Freedom can primarily be characterized by the absence of coercion or constraint. If a person is compelled by the state or the will of another to a course of action or inaction which he would not otherwise have chosen, he is not acting of his own volition and he cannot be said to be truly free. One of the major purposes of the Charter is to protect, within reason, from compulsion or restraint. Coercion includes not only such blatant forms of compulsion as direct commands to act or refrain from acting on pain of sanction, coercion includes indirect forms of control which determine or limit alternative courses of conduct available to others. Freedom in a broad sense embraces both the absence of coercion and constraint, and the right to manifest beliefs and practices. [Emphasis added.]

153 Thus, the test is clearly not that agricultural workers are free to associate as they wish. If that were the case and no associational activity took place, then the conclusion would be that they had the opportunity and chose not to. If the analysis incorporates the reality of the labour market, the chilling effect of the lack of freedom becomes manifest.

154 I agree with my colleague that Professor Fudge is correct when she states in her affidavit that “[w]hile the existence of labour relations legislation is no guarantee that a particular group of workers will be able to join a trade union and engage in collective bargaining with their employer, the absence of such legislative protection virtually guarantees that workers will not enjoy these rights and freedoms”. In that light, it is fair to characterize the freedom of association of agricultural workers under the LRA as being but a hollow right because, as stated by Arthurs, supra, at para. 431, and cited with approval by my colleague, the freedom to organize would amount “to no more than the freedom to suffer serious adverse legal and economic consequences”.

155 As stated by Cory and Iacobucci JJ. in Delisle, supra, at para. 68:

The ability of employees to form and join an employee association is thus crucially linked to their economic and emotional well‑being. Membership in employee groups assists the individual member in a great many ways. Simply to join a trade union is an important exercise of an individual's freedom of expression. It is a group which so often brings to the individual a sense of self‑worth and dignity. An employee association provides a means of openly and frankly discussing work‑related problems without fear of interference or intimidation by the employer. The association provides a means of expressing a collective voice, not only in communicating with the employer, but also in communicating with government, other groups, and the general public. The fundamental importance of the union remains, even though a statute may prohibit the employees from going on strike, or from holding a sit‑in. The freedom of employees to participate in an employee association is basic and essential in our society. A statute whose purpose or effect is to interfere with the formation of employee associations will clearly infringe s. 2(d) of the Charter. [Emphasis added.]

156 The outcome of the exercise of freedom of association is not determinable on an ex ante basis: a meeting by employees could be used as much to organize a picnic as to discuss occupational health and safety issues. The boundary line between permissible and impermissible associational activity is invisible. To use Dickson C.J.’s terminology, in a non-vapid s. 2(d) environment the line would not exist. In an environment where workers do not enjoy protection from unfair labour practices, an employer has no reason to assume that such associational activities will not lead to the forbidden terrain of collective bargaining. The reality is that employers often take anticipatory action against those who undertake organizational activities.

157 In Lavigne, supra, at p. 263, Wilson J. stated that once the positive freedom of association had been established, then it remained to establish whether the appellant has been prevented from forming or joining associations of her or his choosing. In Lavigne, the answer was no. In our case, by contrast, the answer is yes.

158 With respect, I do not share the views of my colleague, as expressed by him in Delisle, at para. 29, that it would be problematic to recognize positive rights because it will force government to take an “all or nothing” approach to the promotion of freedoms. This is because a positive duty to assist excluded groups generally arises when the claimants are in practice unable to exercise a Charter right. In addition, s. 1 of the Charter would allow the government to justify excluding some groups from the application of certain policies. In our case, once the statutory exclusion is removed, the claimants will be in a position to effectively exercise their s. 2(d) rights.

159 However, when the Ontario government repealed the ALRA and enacted the LRA it committed acts which affected the associational rights of agricultural workers in Ontario. As the majority of the Court stated in Vriend, supra, at para. 62:

It might also be possible to say in this case that the deliberate decision to omit sexual orientation from the provisions of the IRPA is an “act” of the Legislature to which the Charter should apply. This argument is strengthened and given a sense of urgency by the considered and specific positive actions taken by the government to ensure that those discriminated against on the grounds of sexual orientation were excluded from the protective procedures of the Human Rights Commission. However, it is not necessary to rely on this position in order to find that the Charter is applicable.

160 The Ontario government, in this case, first ensured that those workers who had initiated associational activities under the protection of the ALRA would no longer pursue them and, secondly, it took “the deliberate decision to omit” persons employed in agriculture from the LRA.

161 In the Alberta Reference, supra, at pp. 376-77, Dickson C.J., in the context of an exclusion clause concerning the right to strike under Alberta’s Hospitals Act, stated that:

The situation with respect to employees of employers who operate approved hospitals under the Hospitals Act is quite different. Prohibiting the right to strike across the board in hospital employment is too drastic a measure for achieving the object of protecting essential services. It is neither obvious nor self‑evident that all bargaining units in hospitals represent workers who provide essential services, or that those who do not provide essential services are “so closely linked” to those who do as to justify similar treatment. As pointed out above, the Freedom of Association Committee of the I.L.O. expressed concern about the overinclusiveness of s. 117.1 of the Labour Relations Act:

132. The Committee notes that this broad exclusion covers kitchen help, janitors, gardeners, etc. . . . Given that this provision is not sufficiently specific as regards the important qualification of “essential employee”, the Committee refers to the principle . . . concerning circumstances in which recourse to strike action may be prohibited. It requests the Government to re‑examine section 117.1 so as to confine the prohibition of strikes to services which are essential in the strict sense of the term. [Emphasis in original.]

162 In our case, the overinclusiveness of the LRA’s impugned provision cannot be justified. Although there may be a rare case where an agricultural operation could justifiably be exempted from the protections of the LRA, the broad exclusion incorporated in the Act, excluding all persons employed in agriculture from all provisions of the Act clearly is, in Dickson C.J.’s words in the above quote, “too drastic a measure”. Such overinclusiveness “could lead to results in certain cases which would defy both rationality and fairness” (Oakes, supra, at p. 142).

163 In Osborne v. Canada (Treasury Board), 1991 CanLII 60 (SCC), [1991] 2 S.C.R. 69, a case concerning a challenge to the federal provisions prohibiting public servants from working for or against candidates or political parties, I concurred with Wilson J.’s reasons that once legislation is found to be over-inclusive, infringes a Charter right and cannot be justified under s. 1, “the Court has no alternative but to strike the legislation down or, if the unconstitutional aspects are severable, to strike it down to the extent of its inconsistency with the Constitution” (pp. 76‑77).

164 The term “underinclusiveness” is often used in s. 15 jurisprudence including cases where, as here, overt exclusion from a statute has taken place. As stated in Vriend, supra, at para. 61, it is the substance not the form of the legislation that matters. However, while the ultimate legal effect may be similar, we should be alert to the impact on the affected parties. There is, after all, a difference between a sign that states “Members Only” from one that states “Agricultural Workers Excluded”.

B. Section 15(1) Analysis

165 The appellants also argue that the LRESLAA and the LRA violate agricultural workers’ equality rights under s. 15(1) of the Charter. Because I have already found that the impugned legislation infringes s. 2(d), I find it necessary to make but a single observation with respect to s. 15(1). I agree with the trial judge that “the central issue to be resolved” under s. 15(1) “is whether [the agricultural workers’] exclusion from the collective bargaining regime constitutes discrimination on an ‘analogous ground’” (p. 209). I disagree, however, that the occupational status of agricultural workers does not satisfy this prong of the s. 15(1) analysis.

166 In Law v. Canada (Minister of Employment and Immigration), 1999 CanLII 675 (SCC), [1999] 1 S.C.R. 497, the Court held, at para. 93, that the determination of whether a ground or confluence of grounds is analogous to those listed in s. 15(1)

is made on the basis of a complete analysis of the purpose of s. 15(1), the nature and situation of the individual or group at issue, and the social, political and legal history of Canadian society’s treatment of the group. A ground or grounds will not be considered analogous under s. 15(1) unless it can be shown that differential treatment premised on the ground or grounds has the potential to bring into play human dignity . . . .

In other words, “[t]o say that a ground of distinction is an analogous ground is merely to identify a type of decision making that is suspect because it often leads to discrimination and denial of substantive equality” (Corbiere v. Canada (Minister of Indian and Northern Affairs), 1999 CanLII 687 (SCC), [1999] 2 S.C.R. 203, at para. 8). A ground need not be immutable to be analogous; it can be based on characteristics that the government has no legitimate interest in expecting claimants to change to receive equal treatment under the law, or, in other words, characteristics that are difficult to change, or changeable only at great cost (Corbiere, supra, at paras. 13-14; see also Vriend, supra, at para. 90). In Egan, supra, I took the position that reliance on grounds amounts to an “indirect means by which to define discrimination” (para. 35 (emphasis in original deleted)), whereas the preferable approach would be to focus on the group adversely affected by the distinction as well as on the nature of the interest affected. I remain convinced that this is the most direct and truthful way of addressing the problem of discrimination. Nonetheless, even under the majority’s current “grounds” approach, there is no reason why an occupational status cannot, in the right circumstances, identify a protected group.

167 First, this Court has repeatedly recognized that employment is a fundamental aspect of an individual’s life and an essential component of identity, personal dignity, self-worth and emotional well-being (see, e.g., McKinley v. BC Tel, [2001] 2 S.C.R. 161, 2001 SCC 38, at para. 53 (quoting Alberta Reference, supra, per Dickson C.J., at p. 368)). Second, though it has had the opportunity to do so, this Court has never declared categorically that a ground of differential treatment based on an occupational status may not be subject to scrutiny under s. 15(1) (see especially Delisle, supra, at para. 44; R. v. Généreux, 1992 CanLII 117 (SCC), [1992] 1 S.C.R. 259, at p. 311). In Delisle, while the majority concluded that RCMP officers did not satisfy this prong of the Law test, it left the door open for the possibility that other occupationally oriented forms of discrimination could fall under the scope of s. 15(1) by limiting its holding to RCMP officers only (see Delisle, supra, at para. 44). In my concurring reasons in that case, I expressed my belief that an occupational status could constitute a suspect marker of discrimination, at para. 8:

[O]ccupation and working life are often important sources of personal identity, and there are various groups of employees made up of people who are generally disadvantaged and vulnerable. Particular types of employment status, therefore, may lead to discrimination in other cases, and should be recognized as analogous grounds when it has been shown that to do so would promote the purposes of s. 15(1) of preventing discrimination and stereotyping and ameliorating the position of those who suffer social and political disadvantage and prejudice.

Legal commentators have also embraced the notion that occupational distinctions between certain groups can be subject to Charter scrutiny (see, e.g., D. Gibson, The Law of the Charter: Equality Rights (1990), at p. 257; D. Pothier, “Connecting Grounds of Discrimination to Real People’s Real Experiences” (2001), 13 C.J.W.L. 37, at p. 57).

168 In this case, there is no doubt that agricultural workers, unlike the RCMP officers in Delisle, do generally suffer from disadvantage, and the effect of the distinction is to devalue and marginalize them within Canadian society. Agricultural workers “are among the most economically exploited and politically neutralized individuals in our society” and face “serious obstacles to effective participation in the political process” (D. M. Beatty, Putting the Charter to Work: Designing a Constitutional Labour Code (1987), at p. 89). Indeed, the trial judge clearly found, at p. 216, that

agricultural workers have historically occupied a disadvantaged place in Canadian society and that they continue to do so today. For the purposes of the s. 15 analysis, I have no hesitation in finding on the evidence that agricultural workers are a disadvantaged group. They are poorly paid, face difficult working conditions, have low levels of skill and education, low status and limited employment mobility.

In light of this, I believe it safe to conclude of agricultural workers what Wilson J. concluded of non-citizens in Andrews v. Law Society of British Columbia, 1989 CanLII 2 (SCC), [1989] 1 S.C.R. 143, at p. 152, namely that they “are a group lacking in political power and as such vulnerable to having their interests overlooked and their rights to equal concern and respect violated. They are among ‘those groups in society to whose needs and wishes elected officials have no apparent interest in attending’”. Thus, the critical question here is whether the government has a legitimate interest in expecting agricultural workers to change their employment status to receive equal treatment under the law.

169 In my view, this question must be answered in the negative. Not unlike the off-reserve aboriginal band members faced with the challenge of changing their status to on-reserve band members identified in Corbiere, I believe that agricultural workers, in light of their relative status, low levels of skill and education, and limited employment mobility, can change their occupational status “only at great cost, if at all” (Corbiere, supra, at para. 14). The fact that the agricultural workforce may be highly transient only reflects the unstable nature of the industry, and does not change the basic point that the workers lack other employment options; indeed, many of the seasonal workers are students and the unemployed. In my view, it is abundantly clear that agricultural workers do not enjoy the same “labour market flexibility” as RCMP officers (Delisle, supra, at para. 44) or other more advantaged professionals, and I see no reason to disturb the trial judge’s considered findings of fact regarding the predicament of agricultural workers.

170 Accordingly, I find that the occupational status of agricultural workers constitutes an analogous ground. I note that in arriving at this conclusion, I make no findings about “occupational status” generally as a suspect marker of discrimination under s. 15(1).

C. Section 1 Analysis

171 The role of s. 1 in the Charter was first fully examined by this Court in Oakes, supra. In R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713, at p. 768, Dickson C.J. summarized the steps in the analysis:

Two requirements must be satisfied to establish that a limit is reasonable and demonstrably justified in a free and democratic society. First, the legislative objective which the limitation is designed to promote must be of sufficient importance to warrant overriding a constitutional right. It must bear on a “pressing and substantial concern”. Second, the means chosen to attain those objectives must be proportional or appropriate to the ends. The proportionality requirement, in turn, normally has three aspects: the limiting measures must be carefully designed, or rationally connected, to the objective; they must impair the right as little as possible; and their effects must not so severely trench on individual or group rights that the legislative objective, albeit important, is nevertheless outweighed by the abridgment of rights.

As with all Charter analysis, a contextual approach is to be followed.

(1) Sufficiently Important Objective

172 Labour statutes, such as the LRA, fulfill important objectives in our society. Harmonious relations between management and labour have an impact not only on the economic relations between the parties, but also on social welfare as a whole. However, for the purposes of our analysis, our focus is on s. 3(b) of the LRA.

173 As indicated earlier in my reasons, I have concluded that s. 3(b) breaches the Charter. This, however, does not necessarily end the s. 1 inquiry; “a legislative provision whose purpose infringes the Charter may nonetheless be found to have an objective that is sufficiently important to justify overriding a Charter freedom” (Delisle, supra, at para. 112, per Cory and Iacobucci JJ.).

174 The respondents argued that there were two factors justifying the passage of s. 3(b). First, that Ontario agriculture has unique characteristics as a result of which it is incompatible with legislated collective bargaining, and second that the LRA’s purposes could not be realized in the agricultural sector.

175 Neither of these arguments are, in my opinion, persuasive. First, it is difficult for me to believe that the production of eggs or mushrooms, let alone all other agricultural products, in Ontario is truly “unique”. Secondly, it is also difficult to accept that none of the LRA’s purposes, which speak to the basic characteristics required for the operation of a modern business, could be realized in the agricultural sector.

176 The LRA also added a new s. 2, titled “Purposes” that states:

2. The following are the purposes of the Act:

3. To facilitate collective bargaining between employers and trade unions that are the freely‑designated representatives of the employees.

4. To recognize the importance of workplace parties adapting to change.

5. To promote flexibility, productivity and employee involvement in the workplace.

6. To encourage communication between employers and employees in the workplace.

7. To recognize the importance of economic growth as the foundation for mutually beneficial relations amongst employers, employees and trade unions.

8. To encourage co‑operative participation of employers and trade unions in resolving workplace issues.

9. To promote the expeditious resolution of workplace disputes.

177 Of the seven purposes of the Act, only one makes explicit reference to collective bargaining. The other six sections refer to the importance of “adapting to change”, the promotion of “flexibility, productivity and employee involvement in the workplace”, the encouragement of “communication between employers and employees in the workplace”, the resolution of “workplace issues”, and the promotion of the “expeditious resolution of workplace disputes”.

178 It is worth reiterating that the sentiments contained in these expressions of intent are not confined to matters pertaining to collective bargaining. They cover issues related to change, employee involvement, communications between management and labour, and the resolution, without limitation, of workplace issues. Thus, the expressions of intent in s. 2 of the LRA are consistent with a policy focused on the importance of ensuring that the productivity of the labour force would continuously improve with the result that the economy would advance towards strengthening its relative competitiveness. This is an important factor in a globalized economic environment, especially given that old-fashioned economic protectionism is neither affordable nor permissible under international trade rules.

179 It would be difficult to argue that such expressions of intent would not be applicable to the agricultural sector. At the very least they would apply to factory-like enterprises. The question also arises whether, if the government of Ontario has concerns about the economic well-being of the agricultural sector, be it its corporative sector or the one represented by “family farms”, when excluding the sector as a whole from the provisions of the Act did it also intend to exclude from this sector the impact of the expressions of intent, essentially comprising a good management wish list, expressed in s. 2 of the statute? The statutory silence on this issue points towards a policy inconsistency to which the legislature, while entitled, ought to have paid effective attention for the purposes of protecting its statutory initiatives from Charter review.

180 In short, the respondents’ argument boils down to this: (a) agriculture is economically vulnerable and barely profitable if at all; (b) the government must ensure that costs are kept down and to this end it has decided to exclude persons employed in agriculture from the LRA, regardless of the size, nature or profitability of the enterprise; (c) the government does not believe that any of the indicia of good management practice listed in s. 2 of the LRA, whose application will assist an enterprise cope with change while enhancing its productivity, and thus enhance profitability, are necessary in any aspect of agricultural production no matter how industrialized.

181 The respondents are asking us to countenance, without enunciating a constitutionally valid reason, a breach of a Charter-guaranteed fundamental right on grounds which appear to be, at least in part, based on a policy geared to enhance the economic well-being of private enterprises. This we cannot do.

182 The government is entitled to provide financial and other support to agricultural operations, including family farms. What is not open for the government to do is to do so at the expense of the Charter rights of those who are employed in such activities, if such a policy choice cannot be demonstrably justified. This they have failed to do.

(2) Proportionality

183 In light of my colleague’s conclusion that the legislation passes the first branch of the Oakes test, I will address the question of whether the impugned measures meet the proportionality branch of this test. The absoluteness of the exclusion clause, barring all persons employed in agriculture from all components of the LRA, speaks to the lack of proportionality between the perceived ills to be avoided and their remedy. For example, are there no situations in Ontario’s agricultural sector where workers should not be so absolutely barred?

(a) Rational Connection

184 At this stage of the proportionality analysis, the respondents must show, on the basis of reason or logic, a causal connection between the objective of protecting the well-being of the agricultural sector in Ontario and the means chosen to secure this objective. While scientific evidence of a causal connection (or of a lack thereof) is relevant at this stage of the s. 1 inquiry, it is not always required (Delisle, supra, at para. 119 (citing RJR‑MacDonald Inc. v. Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 S.C.R. 199, at paras. 153‑54, per McLachlin J.); Ross v. New Brunswick School District No. 15, 1996 CanLII 237 (SCC), [1996] 1 S.C.R. 825, at para. 101).

185 The respondents’ argument is that the stated objective of securing the well-being of the agricultural sector in Ontario can be achieved through the exclusion of persons employed in agriculture from all associational protections contained in the LRA. Excluding the possibility that the impermissible cost-cutting argument is the sole basis for the respondents’ position, the logical connection is not immediately apparent.

186 In fact, the purported means to secure the stated objective, barring all persons employed in agriculture from all the benefits under the LRA, may have the opposite effect. If the LRA’s good labour management principles outlined in s. 2 of the LRA have a basis in fact, then excluding the agricultural labour force from the application of those principles through the comprehensive labour scheme contained in the LRA will lead to an enterprise which would be less well managed, and with lower productivity, than would otherwise be the case. This would impact profitability. As stated by Cory and Iacobucci JJ. in Delisle, supra, at para. 124:

Generally speaking, where this Court has been faced with contradictory evidence of causation for the purpose of the rational connection inquiry, the difficulty has been simply in deciphering whether the evidence supported a causal link. This case raises the somewhat unusual situation that some of the evidence not only does not support a causal link between the legislative objective and the means used to achieve that objective, but it supports precisely the reverse conclusion, namely that the means chosen engender the very mischief sought to be cured. It seems contrary to the purpose of s. 1 of the Charter to find that the state has demonstrably justified its law in circumstances where it is equally probable that the law causes the very social harm it purports to target. [Emphasis added.]

(b) Minimal Impairment

187 As stated by Cory and Iacobucci JJ. in Delisle, supra, at para. 126, “[l]abour relations law is typically an area in which courts have shown the legislature a degree of deference, owing to the complexity and delicacy of the balance sought to be struck by legislation among the interests of labour, management, and the public”. In this regard, they make reference to comments by McIntyre J. in the Alberta Reference, supra, at p. 414:

Labour law, as we have seen, is a fundamentally important as well as an extremely sensitive subject. It is based upon a political and economic compromise between organized labour -- a very powerful socio‑economic force -- on the one hand, and the employers of labour -- an equally powerful socio‑economic force -- on the other. The balance between the two forces is delicate and the public‑at‑large depends for its security and welfare upon the maintenance of that balance. One group concedes certain interests in exchange for concessions from the other. There is clearly no correct balance which may be struck giving permanent satisfaction to the two groups, as well as securing the public interest. The whole process is inherently dynamic and unstable. Care must be taken then in considering whether constitutional protection should be given to one aspect of this dynamic and evolving process while leaving the others subject to the social pressures of the day.

188 I agree with the argument made by Cory and Iacobucci JJ. in Delisle, supra, at para. 129, when they state that “the complete exclusion of a class of employees from a comprehensive labour relations scheme can hardly be characterized as achieving a delicate balance among the interests of labour and those of management and the Canadian public”. The application of such a blunt measure weakens the case for deference to the legislature.

189 This is further aggravated because those affected by the exclusion are not only vulnerable as employees but are also vulnerable as being members of society at large with low income, little education, scant security or social recognition. As stated by Cory and Iacobucci JJ. in Delisle, at para. 130, the Act:

. . . is not designed to protect a vulnerable group in Canadian society. It is true that the public at large is vulnerable to the harmful effects of a police strike. However, in our view, the general public is not a vulnerable group in the sense understood in this Court's s. 1 jurisprudence: see, e.g., Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927, at p. 995, per Dickson C.J. and Lamer and Wilson JJ.; Ross, supra, at para. 88; Thomson Newspapers, supra, at paras. 88‑90, per Bastarache J. The only vulnerable group at issue in the present analysis is RCMP members themselves. Although clearly police officers are not generally considered a vulnerable group within the overall fabric of Canadian society, they are members of a vulnerable group in a relative sense insofar as they are employees. As mentioned above, Dickson C.J. noted in Slaight Communications, supra, at p. 1051, that legislation which seeks to ameliorate the position of employees falls within “a class of cases in which the governmental objective is that of protection of a particularly vulnerable group, or members thereof”. It follows that legislation whose purpose is to maintain the inherent weakness of employees, such as para. (e) in the present case, is not entitled to deference. Indeed, such legislation should be examined with particular care. [Emphasis added.]

190 The question at this stage of the s. 1 inquiry is whether the exclusion expressed by s. 3(b) of the LRA impairs the appellants’ freedom of association as little as reasonably possible in order to achieve the legislation’s objective. McLachlin J. stated, at para. 160 of RJR‑MacDonald, supra:

The impairment must be “minimal”, that is, the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement . . . . On the other hand, if the government fails to explain why a significantly less intrusive and equally effective measure was not chosen, the law may fail.

191 The essential practical question at this stage is whether the Ontario Legislature could have granted to persons employed in agriculture, either within the LRA itself or through the enactment of a separate statute, some of the basic associational protections contained in the LRA without compromising the stated objective of assisting the well-being of the agricultural sector in Ontario (Delisle, supra, at para. 134).

192 In view of the arguments made above, I am of the opinion that the stated objective of securing the well-being of the agricultural sector in Ontario can be achieved through a legislative mechanism that is less restrictive of free association than the existing complete exclusion of agricultural workers from the LRA. The current law is not carefully tailored to balance the Charter freedoms of persons employed in agriculture in Ontario and the societal interest in harmonious relations in the labour market.

193 While it is important to recognize the important role that family farms play in Ontario agriculture, such a role is not unique to Ontario in the context of Canada’s agricultural experience. Since the passage of the Collective Bargaining Act, 1943, both families and farms have evolved in Canada. Both institutions have experienced, and in many cases continue to experience, significant changes and associated difficulties. Both institutions enjoy, and deserve, strong social support from the community at large.

194 However, it is important not to put forward an argument based on a pastoral image which may no longer reflect current reality. In this regard, it is helpful to recall the words of the British Columbia Labour Relations Board, expressed almost a quarter of a century ago, in the case of South Peace Farms and Oil, Chemical and Atomic Workers International Union, Local No. 9-686, [1977] 1 Can. L.R.B.R. 441. The case was decided two years after British Columbia amended its labour legislation to delete provisions expressly excluding farm workers from obtaining bargaining rights under the Labour Code of British Columbia Act, S.B.C. 1973, c. 122. The Board stated that opposition to the extension of labour legislation to farm workers “has been grounded in an anachronistic image of the ‘family farmer’ which is increasingly less accurate” (p. 450). The Board also remarked that the employer in that case was a “sophisticated, well-run business and, in terms of employee relations, much more analogous to an employer in the industrial sector” (p. 449 (emphasis added)). If true in 1977, how much more so today.

195 The Ontario experience with regard to the existence in the agricultural sector of “sophisticated, well-run business and, in terms of employee relations, much more analogous to an employer in the industrial sector” is not dissimilar to that outlined by the British Columbia Labour Relations Board in 1977. My colleague cites in his reasons the case of Wellington Mushroom Farm, [1980] O.L.R.B. Rep. May 813, a case also cited by the OLRB in Cuddy Chicks, supra, at para. 9, where the majority of the board denied LRA certification to the employees of a mushroom factory. However, the majority agreed with the dissenting Board member that “there [was] no ‘industrial relations basis’ for denying [these] employees the right to bargain collectively” (p. 819; see also G. W. Adams, Canadian Labour Law (2nd ed. (loose-leaf)), at p. 6-49). The comments made by the OLRB in Cuddy Chicks, are also applicable in this context.

196 In this case we are being asked by the respondents, without being presented with credible pressing and substantial reasons, to justify distinguishing workers who sort and pack chicken eggs in a factory-like environment from workers who pack and sort Easter eggs in a factory-like environment. The respondents claim that the former workers should not enjoy rights under statutory labour laws, whereas the latter ones are deserving of protection. I do not agree.

197 There is no obvious connection between the exclusion of agricultural workers from the LRA and a farmer: a city-based corporation could be operating an agricultural entity. There is no reference in the LRA to “farmer(s)” or “family farm(s)”. The sole reference in the Act to a farm is the one contained in the expression “farming in all its branches” in the definition of the term “agriculture”. Nor is there a requirement that an owner, or a relative of the owner, of an agricultural enterprise, be it a family farm or a factory-like operation, be personally involved in the operation. Under the Act, an absentee owner of a large agricultural operation would benefit from the restrictions on the freedom of association of its agricultural employees.

198 Other Ontario statutes contain language which makes explicit reference to family farms. For example, the Corporations Tax Act, R.S.O. 1990, c. C.40, defines at s. 1(2) a “family farm corporation” as “a corporation that is throughout the taxation year a corporation, . . . (c) which carried on the business of farming in Ontario through the employment of a shareholder or a member of his or her family actually engaged in the operation of the farm”. Section 1 of the Junior Farmer Establishment Act, R.S.O. 1990, c. J.2, defines a “family farm” as being “a farm operated by a junior farmer and one or more of a spouse of the junior farmer and any persons related to the junior farmer through blood relationship or adoption”. In s. 1(1) of the Fish and Wildlife Conservation Act, 1997, S.O. 1997, c. 41, a “farmer” is defined as “a person whose chief occupation is farming” and “who is living upon and tilling his or her own land, or land to the possession of which he or she is for the time being entitled”.

199 While not determinative of the issue, these statutes are relevant to the argument regarding the overinclusiveness of the exclusion clause in the LRA in that there are ready examples in Ontario’s legislation which have managed agricultural sectors in a more specific manner than in the LRA.

200 In addition, it is worth citing the provisions in the labour statutes of two provinces which incorporate agricultural exemptions.

201 In New Brunswick, the Industrial Relations Act, R.S.N.B. 1973, c. I-4, s. 1(5) stipulates that entities with four or fewer agricultural employees will not be treated as a bargaining unit:

1(5) For the purposes of this Act,

(a) a unit, where an employee is employed in agriculture, shall comprise five or more employees;

202 In Quebec, s. 21 of the Labour Code, R.S.Q., c. C-27, allows a single employee to form a group for the purpose of certification but it also states that:

Persons employed in the operation of a farm shall not be deemed to be employees for the purposes of this division [Certification of Associations of Employees] unless at least three of such persons are ordinarily and continuously so employed.

203 An important point is made clear in the statutes mentioned in the above paragraphs: a family farm, be it in the form of a corporation or not, requires a farmer on the farm. As stated by Dickson J. in Rathwell v. Rathwell, 1978 CanLII 3 (SCC), [1978] 2 S.C.R. 436, at p. 461, when making reference to a family farming operation, that the owners “worked on and operated all of the land as one farm, a family farm”. At p. 457, he stated:

I do not know what term one might properly apply to the Rathwell properties -- “family farm”, or “farming business”, and with all respect to those of a contrary view, I do not think it matters. In one sense, it was a family farm, in another a business, in another it was a way of life. The property was all operated as one family unit by Mr. and Mrs. Rathwell working together.

204 It is clear that not all rural families own “family farms”. The breadwinners in these families are often employed in agriculture. Their Charter rights do not differ from those of other members of the community.

(c) Deleterious and Salutary Effects

205 At this stage, the words of Cory and Iacobucci JJ. in Delisle, supra, at para. 148, are apt:

Having found that the impugned para. (e) of the definition of “employee” in s. 2 of the PSSRA does not minimally impair the appellant’s freedom of association, it is not necessary to consider the proportionality between the importance of the objective and the deleterious effects of the measure, or between the deleterious and salutary effects. We would note, though, that it is unlikely that the provision would be found proportionate at this stage of the inquiry. The exclusion of RCMP members from the PSSRA’s basic associational protections has few, if any, demonstrable salutary effects which could not be achieved by a lesser exclusion. Its negative effects, on both a symbolic level and a practical level, are severe and cut to the core of the Charter’s s. 2(d) protection.

III. Conclusion

206 I agree that the impugned provision in the LRA should be struck down to the extent of its inconsistency with the Charter with the proviso that I agree with my colleague that the Ontario government be given leeway for 18 months for the enactment of a constitutionally compliant replacement.

207 I would answer the constitutional questions as follows:

1. Does s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Canadian Charter of Rights and Freedoms; or

Yes.

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

In view of the answer above there is no need to answer this question.

2. Does s. 3(b) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Charter; or

Yes.

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

In view of the answer above there is no need to answer this question.

3. If the answer to any part of questions 1 or 2 is in the affirmative, is the limitation nevertheless justified under s. 1 of the Charter?

No.

The following are the reasons delivered by

208 Major J. (dissenting) – In spite of the benefit of the reasons of Justices L’Heureux-Dubé and Bastarache, I am unable, principally for the reasons of Sharpe J. (as he then was) in the Ontario Court (General Division), to agree with their disposition of this appeal. In my view, neither s. 2(d) nor s. 15 of the Canadian Charter of Rights and Freedoms is infringed and I would dismiss the appeal.

I. Section 2(d)

209 As identified by Bastarache J., at para. 13, in order to establish a violation of s. 2(d), the appellants must demonstrate that the impugned legislation has, either in purpose or effect, infringed activities protected by s. 2(d). The appellants did not discharge this burden.

A. Purpose of the Exclusion

210 I agree with Bastarache J.’s analysis and conclusion (at para. 32) with regard to the appellants’ failure to establish the unconstitutionality of the purpose of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A (“LRA”).

B. Effect of the Exclusion

211 In Delisle v. Canada (Deputy Attorney General), 1999 CanLII 649 (SCC), [1999] 2 S.C.R. 989, the majority held (at para. 33):

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.

212 At paras. 24-26, Bastarache J. has articulated three factors to be considered in determining whether such “exceptional circumstances” exist in a particular case, such that a positive obligation is imposed on the state by s. 2(d). I respectfully disagree with his conclusion that a consideration of these factors in the present case leads to a finding of a s. 2(d) violation. Specifically, a consideration of the third factor leads to the opposite conclusion on the facts of this case. The third factor is “whether the state can truly be held accountable for any inability to exercise a fundamental freedom” (para. 26). As noted by Bastarache J. at para. 23, in Haig v. Canada, 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995, even if the appellant had been unable to express his views on Quebec secession, this inability was not caused by his exclusion by the state from the national referendum. Thus, the third factor essentially compels an examination of the causal role of the state in the appellants’ inability to exercise the fundamental freedom. In order for the state to “truly be held accountable”, the appellants must be able to demonstrate by direct evidence or inference that the state is causally responsible for his inability to exercise a fundamental freedom, in that the state “substantially orchestrates, encourages or sustains the violation of fundamental freedoms” (para. 26).

213 As recognized by Bastarache J. at para. 42, workers faced significant difficulties organizing prior to the enactment of the LRA. In light of this historic difficulty, I am unable to agree with the following conclusion of Bastarache J. (at para. 45):

In this context, the effect of s. 3(b) of the LRA is not simply to perpetuate an existing inability to organize, but to exert the precise chilling effect I declined to recognize in Delisle.

214 In my opinion, the appellants have failed to establish that the state is causally responsible for the inability of agriculture workers to exercise a fundamental freedom. I conclude that s. 2(d) does not impose a positive obligation of protection or inclusion on the state in this case. As a result, the effect of the impugned legislation does not infringe agriculture workers’ freedom of association.

II. Section 15

215 I adopt the following conclusion of Sharpe J. in this case ((1997), 1997 CanLII 12345 (ON SC), 155 D.L.R. (4th) 193, at pp. 216-17):

However, with reference to identifying personal characteristics, the evidence before me indicates that agricultural workers are a disparate and heterogenous group. There is nothing in the evidence to indicate that they are identified as a group by any personal trait or characteristic other than that they work in the agricultural sector. The evidence indicates that farm owners and operators also suffer from low wages, and that many have low education levels. The low status and prestige of farm workers is similar to that of other manual labourers. In my view, the evidence shows that the legislative decision to exclude agricultural workers from the collective bargaining regime does not reflect stereotypical assumptions about the personal characteristics of agricultural workers, either individually or as a class. Rather, it is based upon the policy-maker’s perception of the characteristics and circumstances of the agricultural industry. The effects of the legislative exclusion impact the diverse group of individuals who work in that sector of the economy and who are not otherwise identifiable as a group.

While a sub-group of temporary seasonal workers brought to Ontario pursuant to a highly structured federal program may be identifiable by race and the status of non-citizen, I fail to see how their situation advances the applicants’ case. These seasonal foreign workers were not covered by ALRA, they are not subject to LRA, and they would not gain the right to be members of a union or enjoy the right to engage in collective bargaining if this application were successful.

In light of this factual record, in the end, the applicants’ case must turn on whether the economic disadvantage of a group of workers, identified as a group only by the fact that they work in a particular sector of the economy, constitutes an analogous ground within the meaning of s. 15(1). I hardly need state that the wisdom, or lack thereof, from the perspective of labour relations policy, of the decision to exclude agricultural workers from collective bargaining has no bearing on this question.

In my view, the disadvantaged position occupied by agricultural workers is not sufficient to constitute the legislative classification “agricultural workers” as an analogous ground for the purposes of s. 15. Economic disadvantage is often the product of discrimination on an analogous ground, and hence serves as a marker that may indicate the presence of such discrimination. There are, however, many causes of economic disadvantage that do not attract the scrutiny of s. 15, and a showing of economic disadvantage does not, by itself, establish discrimination on an analogous ground within the meaning of s. 15. In my view, the absence of evidence of any traits or characteristics analogous to those enumerated in s. 15 which serve to identify those who make up the group of agricultural workers is fatal to their s. 15 claim.

III. Conclusion

216 I would dismiss the appeal. I would answer the constitutional questions as follows:

1. Does s. 80 of the Labour Relations and Employment Statute Law Amendment Act, 1995, S.O. 1995, c. 1, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Canadian Charter of Rights and Freedoms; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

No.

2. Does s. 3(b) of the Labour Relations Act, 1995, S.O. 1995, c. 1, Sched. A, limit the right of agricultural workers

(a) to freedom of association guaranteed by s. 2(d) of the Charter; or

(b) to equality before and under the law and equal benefit of the law without discrimination as guaranteed by s. 15 of the Charter?

No.

3. If the answer to any part of questions 1 or 2 is in the affirmative, is the limitation nevertheless justified under s. 1 of the Charter?

In view of the answers to questions 1 and 2 there is no need to answer this question.

Appeal allowed with costs, Major J. dissenting.

Solicitors for the appellants: Paliare Roland Rosenberg Rothstein, Toronto.

Solicitor for the respondent the Attorney General for Ontario: The Attorney General for Ontario, Toronto.

Solicitors for the respondent Fleming Chicks: Fasken Martineau DuMoulin, Toronto.

Solicitor for the intervener the Attorney General of Quebec: The Attorney General of Quebec, Montréal.

Solicitor for the intervener the Attorney General for Alberta: The Attorney General for Alberta, Edmonton.

Solicitors for the intervener the Canadian Labour Congress: Sack Goldblatt Mitchell, Toronto.

Solicitors for the intervener the Labour Issues Coordinating Committee: Heenan Blaikie, Toronto.

^* See Erratum [2002] 3 S.C.R. iv