Mckinney v. University of Guelph

A number of critics have interpreted the Court's reasons in Dolphin Delivery as drawing a distinction between the common law and legislation and then suggesting that the common law and private litigation are linked and that legislation and litigation in which government is involved are linked. Having interpreted the decision in this manner, the critics then point out that, if this were correct, the Civil Code of Lower Canada would be subject to Charter review but the bulk of the common law would not. Professor Otis puts the point this way:

(Otis, "The Charter, Private Action and the Supreme Court" (1987), 19 Ottawa L. Rev. 71, at p. 87.)

Others have made the same point: see Slattery, "The Charter's Relevance to Private Litigation: Does Dolphin Deliver?" (1987), 32 McGill L.J. 905, at p. 910; and Howse, "Dolphin Delivery: The Supreme Court and the Public/Private Distinction in Canadian Constitutional Law" (1988), 46 U.T. Fac. L. Rev. 248, at p. 251. Moreover, Professor Slattery submits that given that in much of Canada the application of the common law ultimately depends on explicit provisions in various Reception Acts, it is difficult to see how one can justify excluding that common law from Charter review: see Slattery, supra, at p. 910.

Having pointed to one of the ways in which they feel the common law/statute distinction gives rise to difficulties, a number of critics then proceed to attack the distinction at a more general level. Professor Manwaring, for example, observes that it "seems inconsistent to say that the rules governing secondary picketing in British Columbia can be challenged to the extent that they infringe on freedoms of expression solely because they are found in a statute whereas the more restrictive rules in the other jurisdictions cannot be because the legislatures chose consciously not to legislate": see Manwaring, "Bringing the Common Law to the Bar of Justice: A Comment on the Decision in the Case of Dolphin Delivery Ltd." (1987), 19 Ottawa L. Rev. 413, at p. 444.

Professor Slattery accepts that important distinctions exist between the common law and legislation but emphasizes that "these differences are irrelevant to the question of the Charter's application to private relations": see Slattery, supra, at p. 917. He goes on to ask:

And in a passage that captures the essence of much of the criticism directed at the common law/statute distinction, Howse suggests that "McIntyre J.'s identification of common law rules with private ordering and his definition of government action in terms of statute and government activity pursuant to statute represent a formalistic approach to the constitutionally relevant meaning of government action": see Howse, supra, at p. 251.

In my view, this criticism is based on a misinterpretation of the judgment in Dolphin Delivery. I cannot find that McIntyre J. identified the common law with private litigation and legislation with litigation in which government is involved. He in fact made it clear that the crucial element was action by the executive or administrative branch of government based on either legislation which violates the Charter or a common law principle which results in a violation of the Charter. He states very clearly, in my opinion, in the passage I have underlined from p. 599 of his reasons that the Charter applies to the common law, whether in public or private litigation, provided the government has acted upon it. Obviously, it follows from his analysis that while legislation (the act of the legislature) can be subject to Charter review regardless of any executive or administrative action being based upon it, the common law will not be subject to Charter review absent any government action based upon it. This is a necessary conclusion from his view that s. 32(1) requires either a legislative act (legislation broadly construed) or an act of the executive or administrative branch of government based on a common law principle which results in a violation of the Charter. This is the consequence, he states, of the Charter's being made applicable in s. 32(1) to legislatures and governments.

I agree with the commentators that one of the consequences of Dolphin Delivery's refusal to apply the Charter to the common law absent government action is that the Charter will have a broader application in Quebec than in the other provinces. However, it seems inescapable that all legislation including the Civil Code of Quebec is subject to Charter review under s. 32(1) . I see no basis on which the Civil Code can be distinguished for this purpose from other legislation. One might speculate as to whether Parliament overlooked this problem when it enacted s. 32(1), particularly if Professor Hogg is correct that the legislative history supports the view that Parliament did not intend the Charter to apply to private action: see Hogg, Constitutional Law of Canada (2nd ed. 1985), at pp. 23‑24. The necessary result of this, it seems to me, is that government action of some sort is a pre‑requisite for Charter review of common law principles.

The real issue, it seems to me, is whether the Court was correct in concluding that on the wording of s. 32(1) of the Charter government involvement of some kind was required in order to trigger Charter scrutiny. I propose to return to this later.

The second major criticism of McIntyre J.'s judgment, i.e., that it is a mistake not to treat court orders as government action, is particularly challenging because it raises complex questions concerning the very nature of government action. In discussing this aspect of the decision many of Dolphin Delivery's critics have been quick to point to a seeming contradiction in McIntyre J.'s reasoning. For example, Professor Manwaring, supra, states at p. 438:

Other critics have gone on to make at least three points concerning McIntyre J.'s observations about court orders and the apparent tension in his reasoning. First, several writers have stressed that various sections of the Charter make clear that there are instances in which the Charter applies to courts. Professor Hogg, for example, states that ss. 11 , 12 , 13 , 14 and 19 of the Charter obviously apply to the courts: see Hogg, "The Dolphin Delivery Case: The Application of the Charter to Private Action" (1986-87), 51 Sask. L. Rev. 273, at p. 275; see also Howse, supra, at p. 251. Professor Hogg notes that courts in this country have been established or continued by statute and that "their powers to grant injunctions and make other orders are granted (or continued) by statute". Given that other statutory tribunals will have to comply with the Charter, he asks, "Why not the courts?": see Hogg, supra, at p. 275.

A second and more sweeping line of attack suggests that McIntyre J.'s analysis of s. 32(1) is simply incompatible with a robust understanding of s. 52 of the Constitution Act, 1982 . Professor Beatty puts the argument this way:

(Beatty, "Constitutional Conceits: The Coercive Authority of Courts" (1987), 37 U.T.L.J. 183, at p. 187.)

Professor Slattery argues that courts must be seen as a branch of government: courts "act in the name of the community as a whole, as symbolized by the Crown, and derive their authority from that fact. In this respect they represent the State, even if they function differently than other branches of government" (Slattery, supra, at p. 918). Similarly, Professor Gibson states:

(Gibson, " What did Dolphin Deliver?", in Gérald-A. Beaudoin, ed., Your Clients and the Charter -‑ Liberty and Equality (1987), at p. 83.)

Finally, some critics have gone on to articulate a third line of attack on the proposition that court orders are not government action. They have emphasized that it is well accepted in the United States both that court action may constitute government action and that attempts to distinguish courts from government are likely to prove unsuccessful: see, for example, Manwaring, supra, at p. 440. Furthermore, Professor Etherington has observed that many of the academics whose work McIntyre J. found persuasive in Dolphin Delivery conclude that the Charter should not apply to private action and at no time suggest that the Charter does not apply to private litigation: see Etherington, "Retail, Wholesale and Dept. Store Union, Local 580 v. Dolphin Delivery Ltd." (1987), 66 Can. Bar Rev. 818. At page 833, he notes:

To summarize, critics of the proposition that court orders are not government action stress: (i) that various sections of the Charter are obviously applicable to the courts, (ii) that s. 52 of the Constitution Act, 1982 requires that s. 32(1) of the Charter be interpreted in such a way as to bind courts by its provisions, and (iii) that courts represent the state as much as any other branch of government.

Let us return to McIntyre J.'s analysis on this point. The nub of it appears in the passage which I have underlined from p. 600 of his reasons. It states:

Two thoughts underlie this passage, it seems to me. The first is the distinction made by McIntyre J. between the role of the court in litigation as compared with the role of the parties. One of the parties is alleging a Charter violation by the other. The court is bound by the Charter in the sense that it must interpret and apply it to the dispute. But it is, he says, a neutral arbiter in the decision‑making process. The question it has to answer is: has there been a violation of the Charter by either the legislature or the executive or administrative branch of government? The critics say: but it is itself "government" within the meaning of s. 32(1) when it does this. McIntyre J. says no: it is acting in its traditional adjudicative capacity in which it is totally independent of the other branches of government. This must be so, he says, because it could not otherwise perform the function it has been given under the Charter. It cannot be both judge and judged at the same time. How could it, for example, take an unbiased approach to whether the government had violated human rights or whether, if it had, its conduct was justified under s. 1?

This is not to say, as McIntyre J. points out, that the courts are above the law and above the Charter, but simply that in exercising their adjudicative function under the Charter in a dispute between others, they cannot be viewed as "government" and the end product of their decision‑making, the order of the court, as government action for purposes of s. 32(1) .

If, of course, a court as an institution were in its own administration to violate a citizen's human rights, e.g., its employees' freedom of religion or equality rights, it would be just as guilty of a Charter violation as any other institution.

The second thought expressed by McIntyre J. is that, if court orders constitute government action for purposes of s. 32(1) then, since virtually all disputes before the court end in a court order of some kind, all litigation would be subject to Charter scrutiny. McIntyre J. obviously thought that this would be a very convoluted way of making the Charter applicable to private action. Why would s. 32(1) restrict the application of the Charter to legislatures and governments if it was meant to apply to private action as well? Why not simply say so? It is, I believe, also clear from the judgment in Dolphin Delivery that McIntyre J. was concerned that the role of the Human Rights Codes not be pre‑empted by the Charter.

Assuming that my interpretation of the Court's decision in Dolphin Delivery is correct and that the Court did draw a sharp distinction between government and private action for purposes of Charter application, was it justified in so doing?

Professor Slattery has argued that many of the difficulties encountered in Dolphin Delivery flow from the Court's distinction between government and private action. He shares Professor Beatty's view that s. 52 of the Constitution Act, 1982 which states that any law inconsistent with the Constitution "is, to the extent of the inconsistency, of no force or effect" is a clear indication that s. 32(1) was not meant to place limits on the Charter's application. Slattery states at p. 920:

As a result, Professor Slattery suggests that questions of applicability can really only be determined by looking at the individual provisions of the Charter: see Slattery, supra, at p. 922, and Slattery, "Charter of Rights and Freedoms -‑ Does it Bind Private Persons" (1985), 63 Can. Bar Rev. 157, at p. 158.

For his part, Professor Gibson has consistently argued that the only sensible interpretation of s. 32(1) of the Charter is one that places no restrictions on the range of bodies to which it applies: see "The Charter of Rights and the Private Sector" (1982), 12 Man. L.J. 213; "Distinguishing the Governors from the Governed: The Meaning of "Government" Under Section 32(1) of the Charter" (1983), 13 Man L.J. 505; The Law of the Charter: General Principles (1986), at pp. 85‑118; and "What did Dolphin Deliver?", in Your Clients and the Charter -‑ Liberty and Equality, supra, at pp. 75‑90. He stresses that American jurisprudence and academic commentary has struggled in vain to produce a workable distinction. He observes that in Reitman v. Mulkey, 387 U.S. 369 (1967), at p. 378, the United States Supreme Court described efforts to distinguish between private action and government action as an "impossible task". He too is of the view that the wording of s. 32(1) does not require the Court to read limits into the scope of the Charter's application. Moreover, he submits that "If the Charter is to serve the purpose of striking a satisfactory compromise between the claims of the individual and the claims of the community, its norms must be applied to everyone -‑ public or private -‑ whose actions affect the rights and freedoms of others": see The Law of the Charter: General Principles, supra, at p. 118.

Professor Manwaring has also explored some of the problems raised in American jurisprudence that addresses the state action doctrine and notes that there are American writers who have argued that the public/private distinction is conceptually incoherent: see, for example, the Papers from the University of Pennsylvania Law Review Symposium on The Public/Private Distinction (1982), 130 U. Pa. L. Rev. 1289 to 1608. While he observes that in his view s. 32(1) of the Charter was meant to be a codification of the very state action doctrine that has proven the source of so many intractable problems in the United States, he concludes that "The extent of the doctrinal confusion and the strength of the critique suggest that, in spite of the fact that the reasons for including section 32 in the Charter seem obvious, it is going to prove very difficult to apply the section in practice": see Manwaring, supra, at p. 436.

Some commentators who take the position that the Charter applies to private action as well as government action have suggested that s. 32(1) may simply have been included to make it clear that the Charter binds the Crown. For example, Professor De Montigny notes that one might be tempted to explain the presence of this clause by resorting "to the well‑known and long‑established principle that the Crown, in absence of an express indication to the contrary, is not subject to statutory law, and to thereby contend that without express mention of government in section 32, decisions taken by the executive in the exercise of its prerogative powers could not be reviewed": see "Section 32 and Equality Rights", in Bayefsky and Eberts, eds., Equality Rights and the Canadian Charter of Rights and Freedoms (1985), at p. 568.

In similar vein Professor Gibson notes in The Law of the Charter: General Principles, supra, at pp. 112-13:

In other words, this line of argument suggests that had s. 32 not been included, this Court might well have concluded that at least some of the Crown's activities were not subject to the Charter.

I do not find this line of reasoning persuasive since it seems to me obvious that one of the basic purposes of a constitutional document like the Charter is to bind the Crown. I do not believe therefore that in the absence of s. 32(1) it would have been open to the Court to apply ordinary principles of statutory interpretation when construing the Charter and thereby conclude that the Crown was not bound by its provisions.

Moreover, it seems to me that if the purpose of s. 32(1) was simply to make clear that the Charter applies to activities undertaken by virtue of the Crown's common law powers, the provision would have been drafted in much more precise language and that the term "Crown" or "Her Majesty" would have been used. I do not find convincing the suggestion that the term "government" was employed as a more colloquial way of referring to the Crown.

There are, of course, also commentators who agree that providing a clear outline of the limits on Charter application is a very difficult task, but who nonetheless argue that s. 32(1) of the Charter does impose such limits. The problem with Dolphin Delivery, they suggest, is not that the distinction cannot be drawn, but that the Court did not draw it in a satisfactory way. Howse, for example, puts the point this way (supra, at p. 253):

Professor Otis, for his part, observes that it is "remarkable" that the Court did not elaborate on the "jurisprudential and contextual assumptions" underlying its stance. He suggests that when one puts s. 32(1) in its broader context, it becomes clear that the document as a whole was meant to apply only to government: "Many substantive provisions are textually restricted to government, while others have been arguably construed as such by the Supreme Court of Canada". See Otis, supra, at p. 78. In particular, he points to the following provisions of the Charter: s. 19 , which sets out linguistic rights that are clearly aimed at delineating governments' obligations; s. 15, which refers to equality rights only with respect to "law" and which he feels thereby provides strong textual evidence in support of the proposition that the Charter is only applicable to government; ss. 3 and 4, which set out a citizen's democratic rights and which impose corresponding obligations on government; ss. 11 and 13 , which, he submits, this Court has made clear are restricted to criminal and penal proceedings (see Dubois v. The Queen, [1985] 2 S.C.R. 350); and s. 7 which he points out has been interpreted in Operation Dismantle Inc. v. The Queen, [1985] 1 S.C.R. 441, at p. 490, as concerned with the protection of the individual from direct impingement by government upon his or her life, liberty and security of the person. Professor Otis concludes at p. 84:

Professor Hogg also accepts that one must draw a distinction between the acts of private actors and government action. He observes, however, (supra, at p. 274) that:

Professor Hogg develops this argument at greater length in his Constitutional Law of Canada (2nd ed. 1985), at pp. 670‑78. In particular, he suggests, at pp. 675-76, that s. 32(2) of the Charter, which stipulates that, "notwithstanding" s. 32(1) , s. 15 of the Charter was only to come into force three years after s. 32 came into force, "plainly assumes that s. 15 is effective through s. 32(1) ". This, in his view, is evidence that s. 32(1) was meant to limit the application of the Charter. Moreover, he points out that the legislative history of s. 32 supports the view that the Charter has no applicability to private action. He places particular weight on testimony given by Mr. Jordan in 1981 before the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada, who was at the time senior counsel, Public Law, in the Department of Justice. Mr. Jordan [at p. 48:27] asserted that the Charter "addresses itself only to laws and relationships between the state and individuals", not private relationships. Finally, Professor Hogg expresses his conviction that the American state action doctrine captures "the normal, expected role of a constitution: it establishes and regulates the institutions of government, and it leaves to those institutions the task of ordering the private affairs of the people" (supra, at p. 677).

As McIntyre J. pointed out in Dolphin Delivery, supra, at pp. 593‑97, Professor Hogg is not the only one to argue that there are limits on the Charter's application. Professor Swinton has argued that the Charter is neither designed nor suited to deal with private action: see "Application of the Canadian Charter of Rights and Freedoms ", in Tarnopolsky and Beaudoin, eds., The Canadian Charter of Rights and Freedoms ‑‑ Commentary (1982), at p. 41. She observes that the Charter contemplates no positive obligation on governmental bodies to eliminate private discrimination and suggests that the Charter's purpose is to restrain government action, not to generate legislative action (at pp. 46‑47). And at p. 48, she puts forward yet another textually related argument in favour of the proposition that the Charter is limited in its application:

Professor Swinton also suggests that it is important to bear in mind that the Charter is a less effective way to regulate private action than human rights legislation and was not intended to pre‑empt such legislation. She says at p. 48:

And as McIntyre J. noted in Dolphin Delivery, supra, at p. 597, further support for this view may be found in McLellan and Elman, "To Whom Does the Charter Apply? Some Recent Cases on Section 32" (1986), 24 Alta. L. Rev. 361. These authors are also sympathetic to the argument that human rights legislation provides a more efficient and less costly method by which an individual may seek redress for acts of private discrimination (at p. 367).

Where does this leave us? It seems to me that it leaves us where the Court began pre‑Dolphin Delivery, asking itself what the purpose of the Charter was. Was it aimed at government action? Was the Charter perceived by the draftsmen as the intermediary between the citizen and government only or was it also perceived as the intermediary between citizen and citizen? I remain of the view that it was aimed at government action, both legislative and administrative, and that the provincial and federal human rights legislation was left to function within its proper sphere. I do not doubt that the government/private action distinction will be difficult to make in some circumstances but I also believe that the text of the Charter must be respected.

One particularly convincing textual argument, it seems to me, is the proposition that s. 32(1) must be read in light of s. 33, the so‑called override provision. While I do not propose to analyze the nature of the override provision in any detail, particularly since this Court recently had occasion to consider the provision in Ford v. Quebec (Attorney General), [1988] 2 S.C.R. 712, at pp. 733‑45, I believe that the presence of a provision designed to enable the legislature to override certain sections of the Charter lends considerable weight to an interpretation of s. 32(1) that concludes that the focus of the Charter is government. The presence of s. 33 suggests that those governments that subscribed to the Charter were aware that the document was designed to place constraints on their action and that they were concerned to provide themselves with a way to avoid some of those constraints (i.e., ss. 2 and 7 to 15) should this prove necessary.

It seems to me also that this Court's approach to s. 1 of the Charter has emphasized that Charter interpretation is fundamentally about balancing the rights of the citizen against the legitimate objectives of government. At no point has this Court suggested that a s. 1 analysis, notably the proportionality test set out in R. v. Oakes, [1986] 1 S.C.R. 103, is intended to assist in the resolution of disputes between individuals. Indeed, given this Court's approach to s. 1, I have difficulty in seeing how one could engage in a s. 1 analysis absent government action.

No less revealing, in my view, is the fact that provisions like ss. 3‑4 and 16‑20 of the Charter are clearly aimed at legislatures and governments. While no single section can be said to provide conclusive proof that the Charter must be interpreted as concerned solely with government action, I believe that a reading of the document that is sensitive to the need to provide a coherent and consistent interpretation of all of its provisions leads to the conclusion that the purpose of the Charter was to constrain government action.

It is, of course, true that in limiting what government may do, particularly the legislative branch of government, the Charter may place limits on what citizens are entitled to do. But I do not think that this derivative form of constraint supports the proposition that the Charter's focus is as much on constraining the individual as it is on constraining government. On the contrary, it seems to me that a careful analysis of the text as a whole makes clear that, as far as the individual is concerned, the focus of the document is protection and not constraint. It was designed to provide the citizen with constitutionally protected rights and freedoms which he or she could assert against government if the need arose.

While I am sensitive to the observation of Lamer J. [as he then was] in Re B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486, at p. 508, that "the Minutes of the Proceedings of the Special Joint Committee [on the Constitution], though admissible, and granted somewhat more weight than speeches should not be given too much weight", it seems to me that the testimony before that Committee lends support, however limited, to the proposition that the document's focus is on government action. I note that Mr. Jordan, Senior Counsel, Public Law, in the Department of Justice at the time the Charter was before the Special Joint Committee on the Constitution, told the Committee [at p. 48:28] that he thought "the whole of the Charter is addressing itself to the protection for individuals against acts by the state" and that he would be "very worried if we ended up with a Charter that mixed into that the domain of private infringement of liberties and freedoms". He expressed the view [at p. 48:28] that "private" infringements of this kind were best "left to be dealt with by human rights codes": 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, First Session of the Thirty‑second Parliament, 1980‑81, pp. 48:27, 48:28 (January 29, 1981); see also p. 49:47 (January 30, 1981).

The then Minister of Justice (Mr. Chrétien) observed that the Charter was not intended to provide a solution to all social problems and that room had to be left for both levels of government in this country to enact and amend legislation designed to deal with social problems without constantly having to resort to constitutional amendments: see p. 48:27. Although I do not think that any more weight should be placed on testimony regarding the meaning of the term "government" in s. 32(1) than testimony regarding the meaning of the term "liberty" or "equal", we cannot totally ignore the fact that much of the testimony before the Committee is highly compatible with a textual analysis that concludes that the Charter's purpose is to constrain "government", however that term is best understood.

Finally, while it is my view that the textual argument is in and of itself convincing and that ultimately this is the proper basis on which to rest conclusions about the application of the Charter, it seems to me that Professors Swinton, McLellan and Elman have a point when they suggest that the legislatures which enacted the Charter were of the view that the ordering of relations between private individuals was best left to human rights legislation. The thrust of such legislation was to get many disputes out of the courts and into a setting more conducive to providing constructive solutions to various forms of discrimination. I do not believe that the Charter was intended as an alternate route to human rights legislation for the resolution of allegations of private discrimination.

In summary, I remain committed to the view previously expressed by the Court that the Charter applies to government action. And rather than attempt to define the boundary between government action and private action, it seems to me that the focus of our analysis in the group of appeals currently before us must be on the nature of government action. Whether this process will shed light on the debate about the validity of the government/private action distinction need not concern us. What must concern us is: when is action properly attributed to government and what are the criteria by which that determination is to be made? As Roger Tassé points out, "If the Charter applies to everyone, there is no need to define the scope of the government" but if it applies only to government action, then it is vital to ask the question: "What is meant by the word `government' in this context?" See Tassé, "Application of the Canadian Charter of Rights and Freedoms ", in Beaudoin and Ratushny (eds.) The Canadian Charter of Rights and Freedoms (2nd ed. 1989), at p. 97 and 77 respectively.

My colleague La Forest J. has concluded that the Charter applies only to government in its narrowest sense. He finds support for this view in a particular doctrine of the role of constitutions known as "constitutionalism". According to this doctrine states are a necessary evil. Because of the potential for tyranny and abuse which large states embody, the role of government should be strictly confined. Social and economic ordering should be left to the private sector. The more the state interferes with this private ordering, the more likely it is that the freedom of the people will be curtailed. Thus, the minimal state is an unqualified good. However, even with the minimal state there has to be some mechanism to protect the citizen against the risk of government tyranny and that mechanism is the constitution itself. Hence the concept of constitutional government as protector of the citizens' liberty.

Drawing on this vision of the classical role of states and constitutions my colleague has formulated what I would view as a very narrow test of "government action" under s. 32(1) of the Charter. In his view, only those entities which actually are "government" will fall within the ambit of the Charter. They must be "part of the government apparatus", "part of government", "part of the machinery of government".

I believe that the concept of government as oppressor of the people and the function of government as the enactment of "coercive laws" is no longer valid in Canada, if indeed it ever was. To make my point it is necessary to consider the historical evolution of the state in Canada as well as the evolution of its constitution culminating in the document before us, the Canadian Charter of Rights and Freedoms .

The doctrine of constitutionalism was a driving force behind the creation of the American constitution. The American Bill of Rights was in large measure the product of a revolution. Unhappy with the injustices the Americans perceived were perpetrated against them by the British, the American people were left with a deep distrust of powerful states. The United States Constitution enshrines the belief of the American people that unless the state is strictly controlled it poses a great danger to individual liberty. Its primary focus, articulated in the bulk of its provisions, is against "state action". Canada does not share this history.

This Court has already recognized that while the American jurisprudential record may provide assistance in the adjudication of Charter claims, its utility is limited. In Re B.C. Motor Vehicle Act, supra, we were called upon to determine the scope of s. 7 of the Charter. Naturally, at that early stage of Canadian Charter jurisprudence, the American constitutional tradition was heavily relied upon. Nevertheless, Lamer J., writing for the Court, made it eminently clear that our Courts were not to be unduly influenced by the decisions in United States cases. He said at p. 498:

Although in that case Lamer J. was relying primarily on the structural differences that exist between the Canadian and American constitutions, structural differences are not the sole measure of differentiation. Social, political and historical differences between our two nations also exist. The Charter has to be understood and respected as a uniquely Canadian constitutional document. However, the fact that Canada did not spring into being as a nation through the same process as the United States does not necessarily mean that Canadians do not share the same perception as our neighbours of the proper role of government. We can only discern how Canadians perceive that role by examining how it has developed through our history.

Professor Corry in his report The Growth of Government Activities Since Confederation (Ottawa 1939) has emphasized the fact that regulation has always played a role in the governance of Canadian society and that, apart from a brief interlude during the first half of the nineteenth century, the philosophy of laissez‑faire never enjoyed permanent or widespread acceptance here. He commences his discussion of the growth of government activity with the following observation at p. 1:

Professor Risk in his article "Lawyers, Courts, and the Rise of the Regulatory State" (1984), 9 Dalhousie L.J. 31, makes the same point at pp. 32‑33:

While Canada was struggling to become a self‑sufficient nation the popularity of laissez‑faire in England and in the United States was on the wane. As Professor Corry points out, the needs of a new country required the energies of government to be directed towards development. The primary obligation resting on the state in the years immediately following Confederation was the need to open up the country through the establishment of transportation facilities and the provision of basic services.

Indeed, one of the first priorities of the new federation was to knit the country together by the establishment of transportational connections between the various regions. Dorman points out in A Statutory History of the Steam and Electric Railways of Canada, 1836‑1937, (Ottawa 1938), at p. 7:

While it is beyond the scope of this review to detail the myriad ways in which the state has intervened in the railway sector, suffice it to say that the Canadian government has always played a large part in the creation and control of the railways. As Abbott said in his A Treatise on the Railway Law of Canada (Montreal 1896), at p. 1:

It was during those decades that the Canadian economy greatly expanded. This period has been described by a number of authors as the "wheat boom" since during that time, as Mackintosh wrote in his report Economic Background of Dominion‑Provincial Relations (Toronto 1964), at p. 39:

Whether or not the wheat economy was primarily responsible for the economic growth of the period, there is no dispute about the soundness of the general observation that the time was one of significant growth for Canada. The government of the day, headed by Prime Minister Sir Wilfrid Laurier, believed that it was its duty to involve itself in this process. In a speech delivered in 1903 he said:

As Baggaley noted in his review of the role of the Canadian state (The Emergence of the Regulatory State in Canada, 1867‑1939 (Ottawa 1981)), Laurier's conception of the appropriate role of the Canadian government was not novel. He said at pp. 42-43:

The increase in accessability to all regions of the country was accompanied by increased crop production, increased immigration and the growth of Canadian cities. Business also began to grow, in part due to the creation of new enterprises and in part due to the consolidation or merger of smaller businesses. In short, rapid socio‑economic changes were taking place in the early part of this century and those changes sparked a re‑evaluation of the appropriate role of the state. While historians have not always agreed on the characterization of this era of government interventionism most agree that the so-called "progressive era" marked an increased role for and acceptance of government regulation. A remarkable amount of government regulation both economic and social was introduced in this period.

For instance, pure food laws designed to afford basic protections to consumers were enacted during this period. Sellers were compelled to ensure minimum standards of food purity on pain of penalty. The Inland Revenue Act of 1875, S.C. 1874, c. 8, which made it an offence to knowingly sell any adulterated food or drink, exemplifies this kind of legislation. With the increase in industrialization came more sophisticated laws dealing with the market. Under The Food and Drugs Act, 1920, S.C. 1920, c. 27, for example, officers appointed under the Act were given the power to take samples and have them tested for quality by government analysts. Grading and inspection of products was made compulsory and false or misleading labelling was prohibited. Thus, the thrust of these laws shifted from being pure health measures to a regime aimed at protecting the producer's status in the marketplace by providing government guarantees of the quality of his products.

The provinces enacted measures of a similar nature, particularly in the dairy industry. Initial attempts were aimed at correcting the problem of the selling of tainted or diseased products although, as in the case of the federal sphere, these attempts eventually led to a more regulated regime with the added purpose of protecting markets. See for example: The Milk, Cheese and Butter Act, S.O. 1908, c. 55; The Dairy Association Act, S.Q. 1921, c. 37; and Creameries and Dairies Regulation Act, S.B.C. 1920, c. 23.

Legislative forays were also conducted into the employer/employee relationship. Factories Acts were passed in most provinces dealing with the terms of employment of women and children and with sanitation and safety in the work place. By the 1920s all provinces except Prince Edward Island had workers' compensation legislation. Minimum wages and maximum hours of work were established as well. Initially these protections applied only to women and children. It was not until the depression years that mandatory minimum employment standards were recognized as necessary for most workers.

It was during the First World War, however, that the real boom in government regulation during the first half of this century occurred. A number of agencies were created to deal with the problems that a war economy produces, including: a Food Controller, a Fuel Controller, a Paper Controller, the War Trade Board, the Wheat Board, a Board of Commerce, and a Cost of Living Commissioner. Many of the initiatives were short lived, however, and at the end of the war only the Wheat Board remained.

The movement back to a more moderate level of government intervention, one committed to fostering private sector growth, gained sway in the years immediately following the war. It was not to last long, however. The Canadian stock market crash in 1929 ushered in the era of the Great Depression and a dramatic shift in favour of government involvement in market processes and the maintenance of minimum living standards for the population. Ominously, Prime Minister Bennett announced to the country in 1935:

Perhaps because of the great toll the Depression took, a number of welfare oriented pieces of legislation were enacted in the areas of agriculture, labour relations and unemployment. The new measures were unlike the legislation passed in previous decades in that they endorsed the objectives of redistribution and planning. Government began to regulate both prices and output in the agricultural sector. Licensing was introduced in gasoline sales. Restrictions were placed upon the common law remedies of mortgagees and creditors. Some of the important legislative initiatives of that era included: The Farmers' Creditors Arrangement Act, 1934, S.C. 1934, c. 53; The Natural Products Marketing Act, 1934, S.C. 1934, c. 57; The Dominion Trade and Industry Commission Act, 1935, S.C. 1935, c. 59; The Minimum Wages Act, S.C. 1935, c. 44; The Weekly Rest in Industrial Undertakings Act, S.C. 1935, c. 14; The Limitation of Hours of Work Act, S.C. 1935, c. 63; and The Employment and Social Insurance Act, S.C. 1935, c. 38. These statutes, their provisions and effects are thoroughly explored by McConnell in his article, "The Judicial Review of Prime Minister Bennett's `New Deal'" (1968), 6 Osgoode Hall L.J. 39.

A number of commentators date the birth of the Canadian welfare state to the period immediately following the New Deal. Prior to this period there were few provisions aimed at protecting working people and ensuring a minimum standard of living. Before the First World War public education and public health services were virtually the only measures of this kind in place. It was not until later, however, that other forms of income security were introduced. The old age pension scheme was introduced in 1951 and the Guaranteed Income Supplement in 1966. Two employment related measures were also introduced during this period: unemployment insurance in 1940 and the Canada Pension Plan in 1951. Families also began to receive state support in the form of the family allowance and the child tax credit. The provinces continued to provide social assistance to the particularly needy, continuing a tradition that started with the ancient poor laws. The financing of these programs, however, became a joint effort when the federal government introduced the Canada Assistance Program under which a fifty per cent cost sharing agreement was reached with all the provinces except Quebec. In addition, tax deductions for individual pension plans were introduced under the Income Tax Act, R.S.C. 1952, c. 148, as am.

The new wave of social welfare provisions was not limited to income security measures. During the 1950s and 60s a new form of social protection was added: human rights legislation. The first province to enact a statute dedicated solely to the protection of human rights was Saskatchewan which in 1947 passed The Saskatchewan Bill of Rights Act, 1947, S.S. 1947, c. 35. Other provinces, some of which had enacted legislation dealing with specific forms of discrimination in particular sets of circumstances (e.g., the Ontario The Fair Accommodation Practices Act, 1954, S.O. 1954, c. 28), followed suit. Comprehensive codes providing protection on a more global scale began next starting with Ontario in 1962 (The Ontario Human Rights Code, 1961-62, S.O. 1961‑62, c. 93) and ending with Quebec in 1975 (Charter of Human Rights and Freedoms, S.Q. 1975, c. 6). Three provinces have now enacted specific legislation dealing with the problem of pay inequities based on gender: The Pay Equity Act, 1987, S.O. 1987, c. 34; Pay Equity Act, R.S.P.E.I. 1988, c. P‑2; and The Pay Equity Act, S.M. 1985‑86, c. 21.

Nor was the growth of human rights law the last phase in the increasing involvement of the state in the protection of citizens' welfare. The 1970s in particular saw a period of rapid growth in the number of regulatory statutes on such issues as environmental protection, health and safety, and consumer protection. For instance, at the federal level the Arctic Waters Pollution Prevention Act , R.S.C., 1985, c. A‑12 , the Clean Air Act, R.S.C., 1985, c. C‑32, the Environmental Contaminants Act, R.S.C., 1985, c. E‑12, and the Ocean Dumping Control Act, R.S.C., 1985, c. O‑2, were virtually all passed during the first half of that decade. Similarly, government in the 1970s enacted a number of statutes directed at protecting consumers from dangerous or hazardous products such as: the Hazardous Products Act , R.S.C., 1985, c. H‑3 ; the Motor Vehicle Safety Act, R.S.C., 1985, c. M‑10; and the Radiation Emitting Devices Act , R.S.C., 1985, c. R‑1 .

The increase in state activity has naturally led to a large increase in the size of government. In 1962 The Royal Commission on Government Organization (Ottawa) reported that the federal public service had increased nine fold since the First World War and employed some 214,000 civil servants. No fewer than 89 government departments, crown agencies and corporations are listed in the schedules to the Financial Administration Act , R.S.C., 1985, c. F‑11 .

As well, the diversification of state function has led to the creation of a complex conglomeration of entities which together constitute "government". An examination of the range of entities listed in the Financial Administration Act is instructive. For instance, the long tradition of Crown ownership which began with the canals, the Canadian National Railway and provincial public utilities has been continued and many are listed in the schedules. So too are the subsidiaries which these Crown corporations themselves own. Also included are what the Royal Commission on Financial Management & Accountability ("The Lambert Commission") (Ottawa 1980) called shared enterprises and independent deciding and advisory bodies. The latter, which operate with a marked degree of autonomy from government, are nonetheless still considered to be part of the state, illustrating very well the diversity of bodies now considered by the state itself to be part of its enterprise.

In approaching the question of the scope of application of the Charter, I believe we must address the issue of how this very important document became part of Canadian life. While Canada has existed as a nation for over 100 years, it never seems to have been considered necessary or especially desireable prior to 1982 that the Canadian people be protected by an entrenched bill of rights. It is legitimate to ask: why in 1982?

Many commentators have suggested that the increased power of private groups and institutions has resulted in the violation of human freedoms on a massive scale (Tribe, "Refocusing the "State Action" Inquiry: Separating State Acts From State Actors", in Constitutional Choices (Cambridge 1985); Chemerinsky, "Rethinking State Action" (1985), 80 Nw. U.L. Rev. 503; Bazelon, "Civil Liberties ‑‑ Protecting Old Values in the New Century" (1976), 51 N.Y.U. L. Rev. 505; Nerken, "A New Deal for the Protection of Fourteenth Amendment Rights: Challenging the Doctrinal Bases of the Civil Rights Cases and State Action Theory" (1977), 12 Harv. C.R.‑C.L. L. Rev. 297; and Berle, "Constitutional Limitations on Corporate Activity -‑ Protection of Personal Rights from Invasion Through Economic Power" (1952), 100 U. Pa. L. Rev. 933). They argue that private discrimination is hardly trivial and is just as pernicious as discrimination caused by government. As Professor Chemerinsky, supra, put it at pp. 510‑11:

It is not simply that the accumulation of social, political and legal power in private entities makes possible the commission of human rights violations, it is also that recent evidence tends to suggest that it is within the realm of the "private" that the vast bulk of these injustices occur. As Tribe, supra, has remarked (at p. 246):

Clearly, one of the realities of modern life is that "private" power when left unchecked can and does lead to problems which are incompatible with the Canadian conception of a just society. The increasing pressure for and ultimate enactment of human rights legislation speaks eloquently to this fact. Canadian society has been prepared to embrace and solicit the assistance of the state in respect of a number of social, political and economic problems that have plagued our communities from time to time. The Canadian government has thus not been regarded as a monolith of oppression but rather as having a beneficent and protective role to play. Indeed, as Professor Robson points out in his book The Governors and the Governed (London 1964), at pp. 12-13:

This is not to say, as Professor Slattery has remarked in his article, "A Theory of the Charter" (1987), 25 Osgoode Hall L.J. 701, at p. 729, that the Canadian state has not at times been guilty of discriminatory, oppressive, and otherwise inappropriate behaviour towards its citizens. It would be a gross distortion of this nation's history to advance a purist vision of the Canadian way of life. Accordingly, the federal government, recognizing that we are living in a world which is becoming increasingly preoccupied with the problem of effective safeguards for human freedom -‑ witness the Universal Declaration of Human Rights, G.A. Res. 217 A (III), U.N. Doc. A/810, at 71 (1948), the European Convention on Human Rights and Fundamental Freedoms, 213 UNTS 221 and the International Covenant on Civil and Political Rights, 999 UNTS 171 to which Canada became a signatory in 1976 -‑ enacted first the Canadian Bill of Rights, R.S.C., 1985, App. III, in 1960 and then the Canadian Charter of Rights and Freedoms in 1982, the latter having constitutional status. The values reflected in the Charter were to be the foundation of all laws, part of the "supreme law of Canada" against which the constitutionality of all other laws was to be measured.

Several observations may be made with respect to the role of the Canadian state based on this brief historical review.

First, government regulation and intervention has long been part of the political, social and economic culture of Canada though its extent has varied during different periods in our history. The focus of intervention has also changed from time to time in response to different needs. In spite, however, of these fluctuations, it seems to be generally accepted by our historians that the political philosophy of laissez‑faire has not been embraced to any substantial degree in Canada.

Second, as some historians have noted, the phenomenon of the interventionist state has traditionally been and continues to be a feature of Canadian political life. Government participation and control has persisted irrespective of the particular government in power. Thus, as Professor McConnell concludes at p. 222 of his article "Some Comparisons of the Roosevelt and Bennett `New Deals'" (1971), 9 Osgoode Hall L.J. 221:

Third, the interventionist activities of the Canadian state have taken many forms. As noted by Priest, Stanbury and Thompson, ("On the Definition of Economic Regulation", in Stanbury (ed.), Government Regulation: Scope, Growth, Process (Montreal 1980)), policy instruments may take the form of "Moral suasion, exhortation or negotiation", direct expenditures, taxation, tax expenditures and public ownership. All of these measures and probably others are available in order to further the objectives of the state and the Canadian government has utilized many if not all of them at some time or other. It has, for example, engaged in a government‑owned industry in some sectors while merely imposing tariffs in others.

I believe that this historical review demonstrates that Canadians have a somewhat different attitude towards government and its role from our U.S. neighbours. Canadians recognize that government has traditionally had and continues to have an important role to play in the creation and preservation of a just Canadian society. The state has been looked to and has responded to demands that Canadians be guaranteed adequate health care, access to education and a minimum level of financial security to name but a few examples. It is, in my view, untenable to suggest that freedom is co‑extensive with the absence of government. Experience shows the contrary, that freedom has often required the intervention and protection of government against private action.

Finally, it is, I think, true to say that while government intervention has traditionally been acceptable to Canadians, the state has never assumed sole responsibility for economic and social welfare matters. There has always been and continues to be a broad sphere of purely private activity in Canada.

All of these observations lead, in my opinion, to the conclusion that a concept of minimal state intervention should not be relied on to justify a restrictive interpretation of "government" or "government action". Governments act today through many different instrumentalities depending upon their suitability for attaining the objectives governments seek to attain. The realities of the modern state place government in many different roles vis‑à‑vis its citizens, some of which cannot be effected, or cannot be best and most efficiently effected, directly by the apparatus of government itself. We should not place form ahead of substance and permit the provisions of the Charter to be circumvented by the simple expedient of creating a separate entity and having it perform the role. We must, in my opinion, examine the nature of the relationship between that entity and government in order to decide whether when it acts it truly is "government" which is acting. We must, as I suggested at the outset, identify those criteria which are relevant to that determination so that they may be applied in a principled way.

In Hunter v. Southam Inc., [1984] 2 S.C.R. 145, Dickson J., as he then was, emphasized at p. 156, that it was important to engage in a broad purposive analysis of the Charter's provisions. And in R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 344, he stressed that interpretations of the Charter's provisions should be generous rather than legalistic. In deciding what kind of criteria are relevant in interpreting the term "government" in s. 32 of the Charter, we should therefore adopt a purposive approach. We should ask ourselves the question: why does the Charter constrain the activities of government?

It seems to me that a historical review of the growth of the Canadian state makes clear that those who enacted the Charter were concerned to provide some protection for individual freedom and personal autonomy in the face of government's expanding role. I do not think they intended to do this by carving out or preserving "private" spheres of activity. I believe, however, that they considered it crucial to establish norms by which government would be constrained in performing the many roles it has assumed and will no doubt continue to assume. They sought to do this by setting out basic constitutional norms rooted in a concern for individual dignity and autonomy which government should be compelled to respect when structuring important aspects of citizens' lives. The purpose of the Charter then, it seems to me, is to ensure that government action that affects the citizen satisfies these basic constitutional norms. I think that Dickson J. put the point well in Hunter, supra, at p. 155, when he made the following observation about the role of a constitution:

In my view, it follows from these propositions that we must take a broad view of the meaning of the term "government", one that is sensitive both to the variety of roles that government has come to play in our society and to the need to ensure that in all of these roles it abides by the constitutional norms set out in the Charter. This means that one must not be quick to assume that a body is not part of government. Consideration of a wide range of factors may well be necessary before one can conclude definitively that a particular entity is not part of government. If this Court is to discharge its responsibility of ensuring that our constitution does provide "unremitting protection of individual rights and liberties" against government action, then it must not take a narrow view of what government action is. To do so is to limit the impact of the Charter and minimize the protection it was intended to provide.

What then are the criteria which will help us to identify the kinds of bodies that the Charter seeks to constrain through the imposition of constitutional norms? At least three tests have been suggested. While none is probably in and of itself determinative, each has something important to say about the nature of government.

The control test poses the question: is the body in question part of the legislative, executive or administrative branches of government and, if not, is it subject to the control of one of these branches of government? When faced with a body that is not itself part of the legislative, executive or administrative branches of government, the control test in turn asks: (a) general questions about the nature and extent of government control over an entity, such as, "does government exercise such significant control over the operation of the institution that the activities of the latter may properly be seen as activities of the former?"; and (b) more specific questions about the entity's activities, such as, "is there a clear nexus between government and the particular impugned activity?"

In my view, we see a very clear application of this approach in the British Columbia Court of Appeal's decision in the related appeal in Douglas/Kwantlen Faculty Assn. v. Douglas College, [1988] 2 W.W.R. 718. In that case, the Court of Appeal stated at p. 721 that "The control exercised by the government over the affairs of the college generally, coupled with actual governmental involvement in the finalization of the collective agreement, permits no other conclusion [than that the college in question is subject to the Charter]". In reaching this conclusion the Court of Appeal first examined the question of general control. It noted that the College was an agent of the Crown, was subject to ministerial control over many aspects of its activities, and had to have its by‑laws approved by a College Board whose members were appointed by the government.

The Court of Appeal then turned to specific questions concerning the nexus between government and the College's contractual relations with its employees. It noted that the executive branch of government had the power to appoint a Commissioner whose task it was to monitor compensation plans and to investigate arrangements by public sector employers. The Compensation Stabilization Act, S.B.C. 1982, c. 32, gave this Commissioner extensive power to approve or disapprove the terms of collective agreements between the parties. The Court of Appeal was of the view that "In these circumstances the collective agreement must be regarded as the result of an action of the executive or administrative arm of government" (at p. 723).

The general questions the control test requires to be asked in the case of an entity not clearly part of the legislative, executive or administrative branches of government are, in my view, quite apposite. The approach seeks to ascertain whether there is a link between that which one knows is government (i.e., the executive, legislative and administrative branches) and that which one is not sure is government by focussing on whether the former exercises general control over the latter. The challenge under this part of the approach, of course, is to ascertain what are relevant forms of control. While I do not think that one can come up with an exhaustive list of relevant forms of control or that any one form of control will necessarily prove determinative, it does seem to me that the Court of Appeal in Douglas College focussed on the kind of considerations one should bear in mind, viz. whether an actor that is clearly part of a branch of government controls aspects of the entity's activity through input into its policy formulation process, through the approval of the by‑laws or rules that determine how that entity is to carry out its mandate, through the allocation of funding used to implement its objectives, or through the appointment of the personnel that run the entity. These forms of relatively direct control will provide strong indicia that an entity is part of government.

More problematic, in my view, is the second limb of the control test: namely, the search for a specific nexus between government and the impugned act. In many instances, it may be that the relevant branch of government does not exercise control over the entity's activities in as direct a way as in the Douglas College case, but that the entity is nonetheless a governmental actor. One need only think of those bodies that are created by statute, that depend heavily on government funding and that receive broad policy directives concerning their overall mandate from one of the branches of government, but that are deliberately placed at arm's length and given the freedom to make a wide range of choices about how to implement particular policies. This kind of arrangement is hardly novel, particularly in areas where ministers and government departments do not wish to be involved in complex and politically sensitive decisions concerning the allocation of government funds or the specific application of particular policies. Decisions of these kinds often require choosing between irreconcilable demands, and governments have therefore frequently found it prudent to create agencies or tribunals that can make these decisions free from political pressure. Thus, even although such arm's length organizations have often been created with a view to performing tasks that a government department had previously performed or might otherwise have performed, one cannot necessarily point to a nexus between the government and the arm's length organization's day‑to‑day activities.

In my view, it is therefore far from obvious that a body should automatically be deemed to be non‑governmental simply because one cannot point to a specific nexus of the kind seen in Douglas College. To conclude that bodies that are in an arm's length relationship with the executive or administrative branches of government are automatically non‑governmental would mean that a wide range of entities that are created but not controlled by the legislative branch of government would escape Charter review. This would hardly provide the kind of "unremitting protection" of rights and liberties that the Charter was meant to secure.

In other words, the problem with a restrictive application of the control test is that it risks leaving open to government the option to delegate wide powers to arm's length agencies and then to insulate those bodies from Charter review by limiting government involvement in those bodies' day‑to‑day decision-making processes. An unduly restrictive version of the control test would thereby leave it open to government to exclude significant areas of activity from Charter review.

I note that Mr. Roger Tassé has observed, "There has been a tremendous increase in subordinate legislation over the course of the past 25 years. Government by way of regulation is much more commonplace today than is government by conventional legislation": see "Application of the Canadian Charter of Rights and Freedoms ", supra, at p. 73. Mr. Tassé goes on to identify the very concern that I have just raised when he states at p. 72:

In my view, these comments are equally applicable to arm's length bodies that are subject to general governmental control.

It seems to me therefore that the control test has something valuable to say at a general level. The presence of general government control will amount to an important indicium that one is faced with government action although it will not necessarily be conclusive. One can, of course, conceive of entities that are subject to government regulation and that are therefore subject to control but that are in no sense part of government, e.g., private corporations that are subject to government regulation. The evidence that one is dealing with government action will, of course, be even stronger if one can point to a direct nexus between government and the activity in question. But I do not think that the specific questions the control test poses about the presence of such a nexus are in any sense necessary conditions for a finding that there is government action. I am quite prepared to accept that, even in the absence of such a nexus, there may be sufficient government control to enable one to conclude that government action is in issue.

A second test that has been proposed asks whether the performance of a given activity is a "government function". It seems to me that this is the kind of test that the Ontario Court of Appeal applied in this appeal when it asked itself whether a university performs a government function. In the Ontario Court of Appeal's view universities do not perform a government function even although they provide a public service for which they receive significant government funding. But the Court of Appeal felt that a body like a municipality would be subject to the Charter because it performs what the Court of Appeal viewed as quintessentially governmental functions, including the enactment of laws of general application. The Court of Appeal observed (see: McKinney v. University of Guelph (1987), 63 O.R. (2d) 1, at p. 24):

In my view, there are at least three problems with the Ontario Court of Appeal's "functional" approach. First, it seems to me that the particular version of this approach advocated by the Ontario Court of Appeal is based on a rather narrow view of government as the maker and enforcer of laws. At best, this can be but part of any complete picture of the modern Canadian state. I think it clear that over time government has become involved in many areas through the creation of bodies that do not simply enact laws (and may not enact laws at all) but that provide a wide range of services and support (financial or otherwise) to the citizen. There is therefore a real danger that the Ontario Court of Appeal has narrowly circumscribed government's "function" in a way which does not accord with twentieth century reality.

Second, even if one were to operate with a somewhat more expansive concept of a government's "function", this approach would risk excluding from Charter review many actions of the legislative, executive or administrative branches of government that might not necessarily be seen as part of a government's "function": for example, entering into employment agreements with civil servants or entering into contracts for supplies with outside bodies. This result would hardly be compatible with a purposive interpretation of s. 32(1) of the Charter, a provision which states that the Charter applies to "all matters within the authority" of Parliament.

Third, and most importantly, it seems to me that a functional approach risks assuming that government is static, something which the historical review that I have presented reveals is far from the case. If we have learned anything from the widespread criticism of the private/ government distinction and the remarkable evolution of government in the last century, it must surely be that government's functions are not finite. Government has become involved in an ever‑widening range of activities. Moreover, it is likely both to move into new areas and to move out of areas in which it no longer feels it should be involved. Governments' functions are constantly evolving even although there may be some core group of activities that most governments have engaged in most of the time. Any test that focusses solely on these core activities, or that limits itself to the activities that a given government is engaged in at a particular point in time, will be of little use in dealing with hard cases in which government has assumed a new area of involvement.

In other words, it is a mistake to think that one can identify the key function(s) that is (are) determinative of what is government. In my view, it is hardly surprising that in the course of conducting a thorough analysis of a variety of bodies that one might consider part of government, Mr. Tassé concludes that "There are no clear and generally accepted criteria for determining when a function is properly judged to be governmental" (Tassé, supra, at p. 81). A function becomes governmental because a government has decided that it should perform that function, not because the function is inherently a government function. It seems to me that in ignoring this point the functional approach risks putting the cart before the horse. Moreover, it seems to me that one must recognize that there may be circumstances in which both governmental and non‑governmental bodies fulfil a given function at the same time. In such cases the functional approach may tell us little about the status of any given entity that performs that function.

That much having been said, it does seem to me that the functional approach has something to offer, provided that one does not assume that just because a body is not performing a traditional government function it is not a government actor. The fact that an entity is performing an activity that we have come to accept as being one of the exclusive functions that a given level of government performs may well be a strong indicium that one is faced with a government actor. Indeed, one may conclude that even although there is no direct nexus between government and a given body's activities and that even although there is minimal government control over that body, the entity must nonetheless be viewed as part of government because it performs a function that has traditionally been performed by government.

Ultimately, much will turn on the function with which one is concerned. While there are functions that government has long fulfilled, e.g., the criminal law enforcement process, there are others that may on some occasions be fulfilled by government and on other occasions by other kinds of bodies, e.g., private corporations. There may also be functions that government decides it should no longer perform. And, as I have already suggested, there may be sectors of the economy where government is competing directly with the private sector with respect to the provision of particular services and where it is very difficult to apply a functional approach in order to sort out which players are government actors and which are not. At best, then, the functional approach can only provide tentative answers to the question whether one is dealing with government. But the approach may nonetheless point to important considerations that should be taken into account in any analysis of the status of a given body.

A third approach might centre on the question of whether a given body is a "government entity". This approach focusses on the question whether an entity performs a task pursuant to statutory authority and whether it performs that task on behalf of government in furtherance of a government purpose. In my view, this approach captures considerations which neither the control test nor the government function test address, considerations that may well enable us to ascertain whether government is in fact taking on new roles or fulfilling old roles through the creation of new institutional arrangements.

While I am not aware of a decision based on this approach to the interpretation of s. 32(1) of the Charter, it seems to me that this Court has applied a variation of this test in cases in which it has dealt with the doctrine of Crown immunity. I note, for example, that in R. v. Eldorado Nuclear Ltd., [1983] 2 S.C.R. 551, at pp. 565-66, this Court stated:

While this approach has traditionally been used to determine when an entity's actions are not bound by statutes, it seems to me that it may well be of assistance in identifying bodies whose acts are subject to Charter review.

More precisely, this approach looks at the nature of a body's statutory authority and addresses the possibility that government has delegated power to a subordinate body. It seems to me that this approach may therefore assist one to identify those bodies that are neither subject to extensive government control and that cannot be said to be carrying out a traditional government function, but that may nonetheless be the product of government's decision to take on a new role. By examining whether a body exists to serve a government's objectives in a particular area or acts primarily in its own self‑interest, this approach may also assist one in distinguishing between entities that are in some sense creatures of statute but that cannot be said to form part of government (e.g., privately held corporations incorporated under a Business Corporations Act) and entities that are creatures of statute that do form part of government (e.g., Crown agents).

Thus, this approach would assist in identifying bodies like Eldorado Nuclear Limited as part of government even although the body's "corporate objects clauses and the relevant statutes leave it free to operate without government direction" (per Dickson J. in Eldorado, supra, at p. 573) and even although the body operated within a relatively new area of government activity, i.e., the nuclear industry. As a Crown agent created to address what the government of the day clearly perceived to be a matter of public concern, this body would therefore be required to abide by basic constitutional norms.

In my view, this result accords with common sense. I note that in the course of its extensive study of government management and accountability, the Lambert Commission, supra, at p. 269, observed:

As I have already mentioned, it seems to me self‑evident that the Charter was meant to bind the Crown. I can see no reason why Crown agents should be labelled non‑governmental and thereby exempted from the ambit of the Charter. If we are to ensure that the Charter continues to provide unremitting protection of individual rights and liberties, then it seems to me that the "alternative instrumentalities" that the Lambert Commission identified must be subject to the Charter. I note that Professor Hogg has reached a similar conclusion in Constitutional Law of Canada (2nd ed.), supra, at p. 672, where he observes:

Once again, I do not think that this approach will necessarily produce definitive answers. There might well be entities like charitable organizations that are creatures of statute and that serve the public interest, but which would not be properly viewed as part of government. Nevertheless, it does seem to me that this approach captures an important perspective that must be borne in mind in any inquiry concerning government action, a perspective that is absent from both the control test and the government function test. This is a perspective that can help us to identify some of the more unusual bodies that government creates or becomes intricately involved with in the process of pursuing particular government objectives.

As this review of possible approaches to the identification of government makes clear, I do not think that any one test or approach is a panacea. All have something of value to offer since each provides a somewhat different perspective from which to deal with the question what is government. But each alone risks missing a range of bodies that it seems to me must be viewed as part of government, particularly if one is to ensure that the Charter does in fact provide unremitting protection for individual rights and liberties. It would seem therefore that the only satisfactory approach under s. 32(1) of the Charter is one that is sensitive to the strong points of each of the approaches outlined above.

As a result, I would favour an approach that asks the following questions about entities that are not self‑evidently part of the legislative, executive or administrative branches of government:

Each of these questions is meant to identify aspects of government in its contemporary context. An affirmative answer to one or more of these questions would, to my mind, be a strong indicator that one is dealing with an entity that forms part of government. I hasten to add, however, that an affirmative answer can never be more than an indicator. It will always be open to the parties to explain why the body in question is not part of government. Likewise a negative answer is not conclusive that the entity is not part of government. It will always be open to the parties to explain that there is some other feature of the entity that the questions listed above do not touch upon but which makes it part of government.

We must at all costs be sensitive to the fact that government is a constantly evolving organism. It follows that the kinds of questions we must ask when trying to identify government must also be capable of evolving. It seems to me that the reason why fixed tests designed to identify government inevitably fail is that they assume that government is static, an assumption that is not borne out by an historical and comparative review of governments in this and other countries. As a result, the questions that I have listed above are not carved in stone. Other questions may have to be added to the list as governments enter or withdraw from different fields. The questions I have listed are intended only as practical guidelines to those trying to decide whether a body that is not self‑evidently part of the legislative, executive or administrative branches of government may nonetheless be part of government for purposes of s. 32(1) of the Charter.