R. v. Nova Scotia Pharmaceutical Society, [1992] 2 S.C.R. 606

Nova Scotia Pharmaceutical Society, Pharmacy Association of Nova Scotia, Lawtons Drug Stores Limited, William H. Richardson, Empire Drugstores Limited, Woodlawn Pharmacy Limited, Nolan Pharmacy Limited, Christopher D.A. Nolan, Blackburn Holdings Limited, William G. Wilson, Woodside Pharmacy Limited and Frank Forbes Appellants

v.

Her Majesty The Queen Respondent

and

The Attorney General for Ontario and the Attorney General for Alberta Interveners

and

The Association québécoise des pharmaciens propriétaires, Cumberland Drugs (Merivale) Ltd., Kane's Super Drugmart Corp. Ltd., Les Entreprises Norpharm Inc., Escompte Chez Lafortune Inc., Famili‑Prix Inc., Le Groupe Jean Coutu (P.J.C.) Inc., Groupe Pharmaceutique Focus Inc., Les Magasins Koffler de l'Est Inc., McMahon Essaim Inc., Super Escompte Brouillet Inc., B. Mayrand Inc., Superpharm (Montréal) Ltée, Uniprix Inc., Pierre Bossé, François‑Jean Coutu, Claude Gagnon, Guy Lanoue, Michel Lesieur, Guy‑Marie Papillon and Jean‑Guy Prud'Homme Interveners

Indexed as: R. v. Nova Scotia Pharmaceutical Society

File No.: 22473.

1991: December 4; 1992: July 9.

Present: Lamer C.J. and La Forest, L'Heureux‑Dubé, Sopinka, Gonthier, Cory and Iacobucci JJ.

on appeal from the nova scotia supreme court, appeal division

Constitutional law ‑‑ Charter of Rights ‑‑ Fundamental justice ‑‑ Vagueness ‑‑ Conspiracy to prevent or lessen competition unduly ‑‑ Whether word "unduly" in s. 32(1)(c) of Combines Investigation Act so vague as to infringe principles of fundamental justice ‑‑ Canadian Charter of Rights and Freedoms, s. 7 ‑‑ Combines Investigation Act, R.S.C. 1970, c. C‑23, s. 32(1)(c), (1.1).

Constitutional law ‑‑ Charter of Rights ‑‑ Fundamental justice ‑‑ Mens rea ‑‑ Conspiracy to prevent or lessen competition unduly ‑‑ Whether mens rea required by s. 32(1)(c) of Combines Investigation Act inconsistent with principles of fundamental justice ‑‑ Canadian Charter of Rights and Freedoms, s. 7 ‑‑ Combines Investigation Act, R.S.C. 1970, c. C‑23, s. 32(1)(c).

Appeal ‑‑ Supreme Court of Canada ‑‑ Jurisdiction ‑‑ Arguments on appeal ‑‑ Respondent seeking variation of Court of Appeal's reasons on mens rea issue ‑‑ No leave to appeal sought on this issue ‑‑ Issue arising from respondent's notice of intention ‑‑ Whether mens rea issue properly before Court ‑‑ Rules of the Supreme Court of Canada, SOR/ 83‑74, Rule 29(1), (2).

Combines ‑‑ Conspiracy to prevent or lessen competition unduly ‑‑ Mens rea ‑‑ Whether Crown must prove that accused intended to restrict competition unduly ‑‑ Combines Investigation Act, R.S.C. 1970, c. C‑23, s. 32(1)(c) -- Canadian Charter of Rights and Freedoms, s. 7.

Combines ‑‑ Conspiracy to prevent or lessen competition unduly ‑‑ Determination of "undueness" ‑‑ Distinction between questions of fact and questions of law ‑‑ Combines Investigation Act, R.S.C. 1970, c. C‑23, s. 32(1)(c).

The appellants were charged with two counts of conspiracy to prevent or lessen competition unduly, contrary to s. 32(1)(c) of the Combines Investigation Act. Both counts related to the sale and offering for sale of prescription drugs and pharmacists' dispensing services prior to June 16, 1986. They moved for an order quashing the indictment, on the basis that ss. 32(1)(c), 32(1.1) and 32(1.3) of the Act violated ss. 7, 11(a) and 11(d) of the Canadian Charter of Rights and Freedoms and were therefore invalid. The arguments raised revolved essentially on the issues of vagueness and mens rea. The Nova Scotia Supreme Court, Trial Division granted the motion and quashed the indictment. The Appeal Division allowed the Crown's appeal. The main issues raised in this appeal were (1) whether s. 32(1)(c) of the Act infringed s. 7 of the Charter because of vagueness arising from the use of the word "unduly"; and (2) whether s. 32(1)(c) infringed s. 7 by reason of the mens rea required by the offence.

Held: The appeal should be dismissed. Section 32(1)(c) does not violate s. 7 of the Charter.

(1) Vagueness

Vagueness can be raised under s. 7 of the Charter, since it is a principle of fundamental justice that laws may not be too vague. It can also be raised under s. 1 of the Charter in limine, on the basis that an enactment is so vague as not to satisfy the requirement that a limitation on Charter rights be "prescribed by law". Vagueness is also relevant to the "minimal impairment" stage of the Oakes test. Vagueness, when raised under s. 7 or under s. 1 in limine, involves similar considerations and should be considered a single concept. Vagueness as it relates to the "minimal impairment" branch of s. 1 merges with the related concept of "overbreadth".

What is referred to as "overbreadth", whether it stems from the vagueness of a law or from another source, remains no more than an analytical tool to establish a violation of a Charter right. It is always established by comparing the ambit of the provision touching upon a protected right with such concepts as the objectives of the State, the principles of fundamental justice, the proportionality of punishment or the reasonableness of searches and seizures, to name a few. Overbreadth has no autonomous value under the Charter and references to such a doctrine are superfluous.

The "doctrine of vagueness" is founded on the rule of law, particularly on the principles of fair notice to citizens and limitation of enforcement discretion. Fair notice to the citizen comprises a formal aspect ‑‑ an acquaintance with the actual text of a statute ‑‑ and a substantive aspect ‑‑ an understanding that certain conduct is the subject of legal restrictions. The crux of the concern for limitation of enforcement discretion is that a law must not be so devoid of precision in its content that a conviction will automatically flow from the decision to prosecute. The threshold for finding a law vague is relatively high. The factors to be considered include (a) the need for flexibility and the interpretative role of the courts; (b) the impossibility of achieving absolute certainty, a standard of intelligibility being more appropriate, and (c) the possibility that many varying judicial interpretations of a given disposition may exist and perhaps coexist.

The doctrine of vagueness can be summed up in one proposition: a law will be found unconstitutionally vague if it so lacks in precision as not to give sufficient guidance for legal debate -- that is, for reaching a conclusion as to its meaning by reasoned analysis applying legal criteria. The term "legal debate" is not used to express a new standard or one departing from that previously outlined by this Court. It is rather intended to reflect and encompass the same standard and criteria of fair notice and limitation of enforcement discretion viewed in the fuller context of an analysis of the quality and limits of human knowledge and understanding in the operation of the law. The criterion of absence of legal debate relates well to the rule of law principles that form the backbone of our polity. Legal provisions by stating certain propositions outline permissible and impermissible areas, and they also provide some guidance to ascertain the boundaries of these areas. They provide a framework, a guide as to how one may behave, but certainty is only reached in instant cases, where law is actualized by a competent authority. By setting out the boundaries of permissible and non‑permissible conduct, these norms give rise to legal debate. They bear substance, and they allow for a discussion as to their actualization. They therefore limit enforcement discretion by introducing boundaries, and they also sufficiently delineate an area of risk to allow for substantive notice to citizens. No higher requirement as to certainty can be imposed on law in our modern State. The modern State intervenes today in fields where some generality in the enactments is inevitable. The substance of these enactments must remain nonetheless intelligible. The standard of "absence of legal debate" applies to all enactments, irrespective of whether they are civil, criminal, administrative or other. Once the minimal general standard has been met, any further arguments as to the precision of the enactments should be considered at the "minimal impairment" stage of the s. 1 analysis.

Section 32(1)(c) of the Act and its companion interpretative provision s. 32(1.1) do not violate s. 7 of the Charter on grounds of vagueness. Section 32(1)(c) provides that "[e]very one who conspires, combines, agrees or arranges with another person . . . to prevent, or lessen, unduly, competition . . . is guilty of an indictable offence". This section embodies a general standard which represents an intelligible principle, one that carries meaning and that has conceptual force. While the word "unduly" does not have a precise technical meaning, it is a word of common usage which denotes a sense of seriousness. Considering further that s. 32(1)(c) is one of the oldest and most important parts of Canadian public policy in the economic field, and that it mandates a partial rule of reason inquiry into the seriousness of the competitive effects of the agreement, Parliament has sufficiently delineated the area of risk and the terms of debate to meet the constitutional standard. Moreover, s. 32(1)(c) is made even more precise when the content of the inquiry it mandates is considered. The rest of the Act and the case law have outlined a process of examination of market structure and behaviour of the parties to the agreement which eliminates any vagueness that might remain.

In light of the above, the indictment did not infringe s. 11(a) of the Charter.

(2) Mens Rea

The mens rea issue is properly before this Court. Upon filing a notice of intention, the respondent Crown could request a variation of the Court of Appeal judgment on this issue, as long as it ultimately sought to uphold the disposition of the case in the Court of Appeal. Even if the Crown had not filed a notice of intention, the Court would have retained under Rule 29(1) of the Supreme Court Rules complete discretion to treat the whole case as open. A respondent may advance any argument to sustain the judgment below, and he is not limited to the appellant's points of law. This case fell plainly within Rule 29(1), and Rule 29(2), dealing with cross‑appeals, had no application.

The mens rea required by s. 32(1)(c) is not inconsistent with s. 7 of the Charter. While an element of fault must exist before punishment can be justified, a minimum fault requirement with respect to every criminal or regulatory offence satisfies the requirements of s. 7. That fault may be demonstrated by proof of intent, whether subjective or objective, or by proof of negligent conduct, depending on the nature of the offence. Here, the offence set out in s. 32(1)(c) requires the proof of two fault elements: one subjective, the other objective. To satisfy the subjective element of the offence, the Crown must prove that the accused had the intention to enter into the agreement and had knowledge of the terms of that agreement. To satisfy the objective element, the Crown must prove that on an objective view of the evidence adduced the accused intended to lessen competition unduly -- i.e., that the evidence, viewed by a reasonable business person, establishes that the accused was aware or ought to have been aware that the effect of the agreement entered into would be to prevent or lessen competition unduly. Section 32(1)(c) does not therefore violate s. 7 of the Charter.

Cases Cited

Applied: R. v. Wholesale Travel Group Inc., 1991 CanLII 39 (SCC), [1991] 3 S.C.R. 154; considered: Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 CanLII 105 (SCC), [1990] 1 S.C.R. 1123; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927; R. v. Keegstra, 1990 CanLII 24 (SCC), [1990] 3 S.C.R. 697; R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103; Canada (Human Rights Commission) v. Taylor, 1990 CanLII 26 (SCC), [1990] 3 S.C.R. 892; 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; R. v. Butler, 1992 CanLII 124 (SCC), [1992] 1 S.C.R. 452; R. v. Morgentaler, 1988 CanLII 90 (SCC), [1988] 1 S.C.R. 30; referred to: Container Materials Ltd. v. The King, 1942 CanLII 1 (SCC), [1942] S.C.R. 147; R. v. Vaillancourt, 1987 CanLII 2 (SCC), [1987] 2 S.C.R. 636; Howard Smith Paper Mills Ltd. v. The Queen, 1957 CanLII 11 (SCC), [1957] S.C.R. 403; Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486; Knox Contracting Ltd. v. Canada, 1990 CanLII 71 (SCC), [1990] 2 S.C.R. 338; R. v. Hess, 1990 CanLII 89 (SCC), [1990] 2 S.C.R. 906; Atlantic Sugar Refineries Co. v. Attorney General of Canada, 1980 CanLII 226 (SCC), [1980] 2 S.C.R. 644; Aetna Insurance Co. v. The Queen, 1977 CanLII 174 (SCC), [1978] 1 S.C.R. 731; Perka v. The Queen, 1984 CanLII 23 (SCC), [1984] 2 S.C.R. 232; Village of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489 (1982); Broadrick v. Oklahoma, 413 U.S. 601 (1973); R. v. Smith, 1987 CanLII 64 (SCC), [1987] 1 S.C.R. 1045; R. v. Goltz, 1991 CanLII 51 (SCC), [1991] 3 S.C.R. 485; R. v. Zundel (1987), 1987 CanLII 121 (ON CA), 58 O.R. (2d) 129; Maynard v. Cartwright, 486 U.S. 356 (1988); Grayned v. City of Rockford, 408 U.S. 104 (1972); R. v. MacDougall, 1982 CanLII 212 (SCC), [1982] 2 S.C.R. 605; Papachristou v. City of Jacksonville, 405 U.S. 156 (1972); Kolender v. Lawson, 461 U.S. 352 (1983); Smith v. Goguen, 415 U.S. 566 (1974); Eur. Court H. R., Sunday Times case, judgment of 26 April 1979, Series A No. 30; Eur. Court H. R., Malone case, judgment of 2 August 1984, Series A No. 82; Eur. Court H. R., Kruslin case, judgment of 24 April 1990, Series A No. 176‑A; Eur. Court H. R., Huvig case, judgment of 24 April 1990, Series A, No. 176‑B; Eur. Court H. R., case of Silver and others, judgment of 25 March 1983, Series A No. 61; Eur. Court H. R., Barthold case, judgment of 25 March 1985, Series A No. 90; Eur. Court H. R., case of Müller and others, judgment of 24 May 1988, Series A No. 133; Eur. Court H. R., Leander case, judgment of 26 March 1987, Series A No. 116; R. v. Wigglesworth, 1987 CanLII 41 (SCC), [1987] 2 S.C.R. 541; R. v. Shubley, 1990 CanLII 149 (SCC), [1990] 1 S.C.R. 3; Weidman v. Shragge (1912), 1912 CanLII 58 (SCC), 46 S.C.R. 1; Stinson‑Reeb Builders Supply Co. v. The King, 1929 CanLII 83 (SCC), [1929] S.C.R. 276; R. v. Elliott (1905), 1905 CanLII 117 (ON CA), 9 C.C.C. 505; R. v. J. J. Beamish Construction Co. (1967), 1967 CanLII 239 (ON CA), 65 D.L.R. (2d) 260; General Motors of Canada Ltd. v. City National Leasing, 1989 CanLII 133 (SCC), [1989] 1 S.C.R. 641; R. v. J. W. Mills & Son Ltd., 1968 CanLII 1279 (CA EXC), [1968] 2 Ex. C.R. 275; R. v. Canadian Coat and Apron Supply Ltd., 1967 CanLII 1157 (CA EXC), [1967] 2 Ex. C.R. 53; R. v. Anthes Business Forms Ltd. (1975), 1975 CanLII 54 (ON CA), 26 C.C.C. (2d) 349; R. v. Canadian General Electric Co. (1976), 1976 CanLII 756 (ON SC), 34 C.C.C. (2d) 489; Association québécoise des pharmaciens propriétaires v. Canada (Procureur général), [1991] R.J.Q. 205; R. v. Metropolitan Toronto Pharmacists' Association (1984), 1984 CanLII 5940 (ON SC), 3 C.P.R. (3d) 233; R. v. Abitibi Power & Paper Co. (1960), 1960 CanLII 501 (QC CQ), 131 C.C.C. 201; NCAA v. Board of Regents of the University of Oklahoma, 468 U.S. 85 (1984); Federal Trade Commission v. Indiana Federation of Dentists, 476 U.S. 447 (1986); C.J.E.C., Völk v. Établissements J. Vervaecke S.p.r.l., Case 5/69, [1969] E.C.R. 295; C.J.E.C., S.A. Cadillon v. Firma Höss Maschinenbau K.G., Case 1/71, [1971] C.J.E.C. Rep. 351; R. v. McGavin Bakeries Ltd. (1951), 1951 CanLII 250 (AB SCTD), 3 W.W.R. 289; R. v. Northern Electric Co., 1955 CanLII 392 (ON SC), [1955] 3 D.L.R. 449; Hills v. Canada (Attorney General), 1988 CanLII 67 (SCC), [1988] 1 S.C.R. 513.

Statutes and Regulations Cited

1984 Merger Guidelines, 49 Fed. Reg. 26823.

Act for the Prevention and Suppression of Combinations formed in restraint of Trade, S.C. 1889, c. 41.

Canadian Charter of Rights and Freedoms, ss. 1, 2, 7, 8, 11(a), (d), (h).

Civil Code of Lower Canada, art. 1053.

Combines Investigation Act, R.S.C. 1970, c. C‑23 [am. 1974‑75‑76, c. 76, s. 14], ss. 30(2) [rep. & subs. 1986, c. 26, s. 28], 32(1)(c), (1.1), (1.3) [ad. idem, s. 30(3)], (2), (3), (6), 32.01 [ad. idem, s. 31], 51(7) [ad. idem, s. 47], 70 [idem].

Commission notice of 3 September 1986 on agreements of minor importance which do not fall under Article 85(1) of the Treaty establishing the European Economic Community, O.J.E.C., 12 September 1986, No. C 231/2.

Commission Regulation (EEC) No. 1983/83 of 22 June 1983 on the application of Article 85(3) of the Treaty to categories of exclusive distribution agreements, O.J.E.C., 30 June 1983, No. L 173/1, arts. 1, 2.

Competition Act, R.S.C., 1985, c. C‑34 [am. c. 19 (2nd Supp.), s. 19], ss. 34(2) [rep. & subs. idem, s. 28], 45(1)(c), (2), (2.2) [ad. idem, s. 30(3)], 45.1 [ad. idem, s. 31], 79(7) [idem, s. 45], 98 [idem].

Constitution of the United States, First Amendment, Fifth Amendment, Eighth Amendment, Fourteenth Amendment.

Criminal Code, R.S.C., 1985, c. C‑46, ss. 19, 219, 222‑240.

European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 222, Arts. 8(2), 9(2), 10(2), 11(2).

Protocol No. 4 to the European Convention for the Protection of Human Rights and Fundamental Freedoms, securing certain rights and freedoms other than those included in the Convention and in the first Protocol thereto, Europ. T.S. No. 46, Art. 2(3).

Rules of the Supreme Court of Canada, SOR/83‑74, r. 29(1) [rep. & subs. SOR/88‑247, s. 10], (2).

Sherman Act, c. 647, 26 Stat. 209 (1890), s. 1 (codified as amended 15 U.S.C. §§ 1‑7 (1982)).

Treaty establishing the European Economic Community, Art. 85.

Authors Cited

Amselek, Paul. "La teneur indécise du droit" (1991), 107 Rev. dr. publ. 1199.

Areeda, Phillip E. Antitrust Law: An Analysis of Antitrust Principles and Their Application, vol. 7. Boston: Little, Brown & Co., 1987.

Areeda, Phillip E. and Herbert Hovenkamp. Antitrust Law: An Analysis of Antitrust Principles and Their Application (1990 Supplement). Boston: Little, Brown & Co., 1990.

Baudouin, Jean‑Louis. Les obligations, 3rd ed. Cowansville, Qué.: Éditions Yvon Blais Inc., 1989.

Blaau, Loammi C. "The Rechtsstaat Idea Compared with the Rule of Law as a Paradigm for Protecting Rights" (1990), 107 S. Afr. L.J. 76.

Carré de Malberg, R. Contribution à la théorie générale de l'État, t. 1. Paris: Sirey, 1920.

Chevallier, Jacques. "L'État de droit" (1988), 104 Rev. dr. publ. 313.

Chitty on Contracts: General Principles, 25th ed. London: Sweet & Maxwell, 1983.

Colvin, Eric. "Criminal Law and The Rule of Law". In Patrick Fitzgerald, ed., Crime, Justice & Codification: Essays in commemoration of Jacques Fortin. Toronto: Carswell, 1986, 125.

Dunlop, Bruce, David McQueen and Michael Trebilcock. Canadian Competition Policy: A Legal and Economic Analysis. Toronto: Canada Law Book Inc., 1987.

Henry, Jean‑Pierre. "Vers la fin de l'État de droit ?" (1977), 93 Rev. dr. publ. 1207.

Jeffries, Jr., John Calvin. "Legality, Vagueness, and the Construction of Penal Statutes" (1985), 71 Va. L. Rev. 189.

Neumann, Franz. The Rule of Law: Political Theory and the Legal System in Modern Society. Leamington Spa, Warwirckshire: Berg Publishers Ltd., 1986.

Rogerson, Carol. "The Judicial Search for Appropriate Remedies Under the Charter: The Examples of Overbreadth and Vagueness". In Robert J. Sharpe, ed., Charter Litigation. Toronto: Butterworths, 1987, 233.

Stanbury, W. T. Legislation to Control Agreements in Restraint of Trade in Canada: Review of the Historical Record and Proposals for Reform. National Conference on the Centenary of Competition Law and Policy in Canada, October 1989.

Stanbury, W. T., and G. B. Reschenthaler. "Reforming Canadian Competition Policy: Once More unto the Breach" (1981), 5 Can. Bus. L.J. 381.

Tribe, Laurence H. American Constitutional Law, 2nd ed. Mineola, N.Y.: Foundation Press, Inc., 1988.

APPEAL from a judgment of the Nova Scotia Supreme Court, Appeal Division (1991), 1991 CanLII 2553 (NS CA), 102 N.S.R. (2d) 222, 279 A.P.R. 222, 80 D.L.R. (4th) 206, 64 C.C.C. (3d) 129, 36 C.P.R. (3d) 173, 7 C.R.R. (2d) 352, setting aside a judgment of the Trial Division (1990), 1990 CanLII 2604 (NS SC), 98 N.S.R. (2d) 296, 263 A.P.R. 296, 73 D.L.R. (4th) 500, 59 C.C.C. (3d) 30, 32 C.P.R. (3d) 259, allowing the appellants' motion to quash the indictment. Appeal dismissed.

Joel Fichaud, H. E. Wrathall, Q.C., and Catherine Walker, for the appellants.

Michael R. Dambrot, Q.C., Calvin S. Goldman, Q.C., and John S. Tyhurst, for the respondent.

M. Philip Tunley and David B. Butt, for the intervener the Attorney General for Ontario.

Bart Rosborough, for the intervener the Attorney General for Alberta.

Yves Bériault and Madeleine Renaud, for the interveners the Association québécoise des pharmaciens propriétaires et al.

//Gonthier J.//

The judgment of the Court was delivered by

Gonthier J. --

I. Facts and Proceedings

The twelve appellants were indicted on May 31, 1990, with two counts of conspiracy to prevent or lessen competition unduly, contrary to s. 32(1)(c) of the Combines Investigation Act, R.S.C. 1970, c. C-23. Both counts related to the sale and offering for sale of prescription drugs and pharmacists' dispensing services between January 1, 1974 and June 16, 1986, for the first, and between July 1, 1976 and June 16, 1986, for the second. The trial was set to begin in October of 1990.

On August 21, 1990, the appellants made a motion for an order to quash the indictment, on the basis that ss. 32(1)(c), 32(1.1) and 32(1.3) of the Act violated ss. 7, 11(a) and 11(d) of the Canadian Charter of Rights and Freedoms and were therefore invalid. The arguments raised revolved essentially on the issues of vagueness and mens rea. On September 5, 1990, Roscoe J. of the Nova Scotia Supreme Court, Trial Division, allowed the motion and quashed the indictment: (1990), 1990 CanLII 2604 (NS SC), 98 N.S.R. (2d) 296, 263 A.P.R. 296, 73 D.L.R. (4th) 500, 59 C.C.C. (3d) 30, 32 C.P.R. (3d) 259. The respondent appealed to the Nova Scotia Supreme Court, Appeal Division. On April 24, 1991, a unanimous bench (Clarke C.J.N.S., Jones and Hallett JJ.A.) allowed the appeal: (1991), 1991 CanLII 2553 (NS CA), 102 N.S.R. (2d) 222, 279 A.P.R. 222, 80 D.L.R. (4th) 206, 64 C.C.C. (3d) 129, 36 C.P.R. (3d) 173, 7 C.R.R. (2d) 352. A notice of appeal was filed in this Court on May 22, 1991.

II. Relevant Statutory Provisions

Combines Investigation Act

32. (1) Every one who conspires, combines, agrees or arranges with another person

(c) to prevent, or lessen, unduly, competition in the production, manufacture, purchase, barter, sale, storage, rental, transportation or supply of a product, or in the price of insurance upon persons or property,

is guilty of an indictable offence and is liable to imprisonment for five years or a fine of one million dollars or to both.

The Act was amended in 1976 by S.C. 1974-75-76, c. 76, also known as "Stage I" of competition law reform. Section 32(1.1) was then added:

(1.1) For greater certainty, in establishing that a conspiracy, combination, agreement or arrangement is in violation of subsection (1), it shall not be necessary to prove that the conspiracy, combination, agreement or arrangement, if carried into effect, would or would be likely to eliminate, completely or virtually, competition in the market to which it relates or that it was the object of any or all of the parties thereto to eliminate, completely or virtually, competition in that market.

In 1986, in the course of "Stage II" of the reform, S.C. 1986, c. 26, further added s. 32(1.3) to the Act (renamed the Competition Act):

(1.3) For greater certainty, in establishing that a conspiracy, combination, agreement or arrangement is in contravention of subsection (1), it is necessary to prove that the parties thereto intended to and did enter into the conspiracy, combination, agreement or arrangement, but it is not necessary to prove that the parties intended that the conspiracy, combination, agreement or arrangement have an effect set out in subsection (1).

These sections are now respectively known as ss. 45(1)(c), (2) and (2.2) of the Competition Act, R.S.C., 1985, c. C-34.

III. Judgments Below

Nova Scotia Supreme Court, Trial Division

On the mens rea issue, Roscoe J. reviewed the case law and concluded that s. 32(1)(c) of the Act requires the Crown to prove only that the accused intended to enter into an agreement, the effect of which, if carried out, would be to lessen competition, but not that it also intended to prevent or lessen competition unduly. Relying on R. v. Vaillancourt, 1987 CanLII 2 (SCC), [1987] 2 S.C.R. 636, she found that s. 32(1)(c) creates a truly criminal offence, and that the absence of a subjective mens rea requirement with respect to the lessening of competition leaves the possibility that the "morally innocent" be convicted. She therefore concluded that s. 32(1)(c) violates s. 7 of the Charter.

On the vagueness issue, Roscoe J., after having considered the Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 CanLII 105 (SCC), [1990] 1 S.C.R. 1123 (hereinafter the "Prostitution Reference"), examined the case law interpreting the word "unduly" in s. 32(1)(c) of the Act. She was of the opinion that the only test that provided a sufficient degree of certainty to meet the standards of s. 7 was the "virtual elimination of competition" test enunciated by Cartwright J. in Howard Smith Paper Mills Ltd. v. The Queen, 1957 CanLII 11 (SCC), [1957] S.C.R. 403. Since that test had been repealed by the enactment in 1976 of s. 32(1.1), Roscoe J. held that s. 32(1)(c) was too vague and violated s. 7 of the Charter. She also held the indictment too vague, on the basis that the mere repetition of the words of s. 32(1)(c) in the indictment could not give sufficient notice and information to the accused, and deprived them of their right to a full answer and defence under ss. 7, 11(a) and 11(d) of the Charter.

Roscoe J. considered that s. 1 of the Charter could not be applied to cure the violations of s. 7 flowing from the mens rea requirement, following the dictum of Lamer J. (as he then was) in Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486, as well as appellate pronouncements on the issue. She also considered that the vagueness of s. 32(1)(c) could not make it a limit "prescribed by law" within the meaning of s. 1 of the Charter, and thus that this violation of s. 7 could not be saved either. She declared ss. 32(1)(c) and 32(1.1) of the Act invalid and of no force and effect, and quashed the indictment.

Supreme Court of Nova Scotia, Appeal Division (1991), 1991 CanLII 2553 (NS CA), 64 C.C.C. (3d) 129

For the Court, on the mens rea issue, Clarke C.J.N.S. viewed Atlantic Sugar Refineries Co. v. Attorney General of Canada, 1980 CanLII 226 (SCC), [1980] 2 S.C.R. 644, as having effected a change in law. Following the decision of this Court, he held that s. 32(1)(c) requires the Crown to prove that the accused intended unduly to lessen competition. Furthermore, Clarke C.J.N.S. characterized s. 32(1)(c) as a criminal offence, even though it is found in a regulatory statute, citing Knox Contracting Ltd. v. Canada, 1990 CanLII 71 (SCC), [1990] 2 S.C.R. 338, in support. In the light of Vaillancourt and R. v. Hess, 1990 CanLII 89 (SCC), [1990] 2 S.C.R. 906, he held that "where a statutory provision creates an offence and imposes the possibility of imprisonment as a penalty upon conviction, some degree of mens rea must attach to each essential element of the offence if the provision is to comply with s. 7 of the Charter" (p. 146). Given his conclusion on the substance of the section, there was no infringement of s. 7.

Clarke C.J.N.S. did not rule on the validity of s. 32(1.3) of the Act. Since it effected a substantive change in the law by removing the subjective mens rea requirement with respect to the lessening of competition, this section could not have a retroactive effect, and therefore it did not apply to the charges against the accused.

With respect to the vagueness argument, Clarke C.J.N.S. considered that the impugned provision must be assessed in light of the Act as a whole and of the relevant case law. First of all, in ss. 32(2) and 32(3), the Act enumerates subject-matters that will or will not attract the application of s. 32(1). Section 32(1.1) indicates that a virtual elimination of competition is not necessary to constitute the offence.

Secondly, case law has established that the inquiry must focus on the effect of the agreement on competition in the related market. A host of considerations then come into play, constituting a framework for decision. Clarke C.J.N.S. could not find s. 32(1)(c) vague, as it has been given meaning by the courts for a long period of time, leading to convictions and acquittals. The word "unduly" actually benefits the accused; even though it defies precise measurement, it is of common usage, and denotes a sense of seriousness. Parties to an impugned agreement were in the best position to assess the likely effect on competition.

Section 32(1)(c) was held by Clarke C.J.N.S. not to be unconstitutionally vague, and consequently the indictment was held not to violate ss. 7, 11(a) or 11(d) of the Charter. The appeal was allowed.

IV. Issues

The following constitutional questions were stated by the Chief Justice on July 11, 1991:

1. Is s. 32(1)(c) of the Combines Investigation Act, R.S.C. 1970, c. C‑23 (now s. 45(1)(c) of the Competition Act, R.S.C., 1985, c. C‑34) in whole or in part inconsistent with s. 7 of the Canadian Charter of Rights and Freedoms?

2. Is s. 32(1.1) of the Combines Investigation Act, R.S.C. 1970, c. C‑23 (now s. 45(2) of the Competition Act, R.S.C., 1985, c. C‑34) inconsistent with s. 7 of the Canadian Charter of Rights and Freedoms?

3. If the answer to questions 1 or 2 is yes, is the infringement nevertheless justified under s. 1 of the Canadian Charter of Rights and Freedoms?

Given the structure of the arguments presented by the parties, I propose to deal with the various issues as follows:

I. The alleged unconstitutional vagueness of ss. 32(1)(c) and 32(1.1) of the Act and of the indictment;

II. A. The mens rea required by s. 32(1)(c) of the Act; and

B. The constitutionality of the mens rea requirement of s. 32(1)(c) of the Act.

In their notice of appeal, the appellants raised only the vagueness issue as a ground of appeal. By a notice of intention of June 20, 1991, the respondent indicated that it would seek a variation of the appeal judgment on the mens rea issue. The respondent asks this Court to hold that s. 32(1)(c) does not require the Crown to prove intent to lessen competition, and that s. 32(1)(c) nevertheless does not violate the Charter. The appellants contend that the respondent was forbidden to raise the mens rea issue without having obtained prior leave from the Court under Rule 29(2) of the Rules of the Supreme Court of Canada, SOR/83-74, since the respondent is indeed cross-appealing from the appeal judgment.

The respondent through its notice of intention is not in fact launching a cross-appeal from the decision of the Supreme Court of Nova Scotia, Appeal Division. It does not seek to modify the disposition of the case. It only aims at varying the reasons given for that disposition. This case falls plainly within Rule 29(1) of the Rules of the Supreme Court. Rule 29(2), dealing with cross-appeals, has no application here. Upon filing a notice of intention, the respondent could request a variation of the Court of Appeal judgment on the mens rea issue, as long as it ultimately upholds its disposition of the case. Even if the respondent had not filed a notice of intention, the Court would have retained under Rule 29(1) complete discretion to treat the whole case as open, as was done in Perka v. The Queen, 1984 CanLII 23 (SCC), [1984] 2 S.C.R. 232. As Dickson J. (as he then was) wrote at p. 240, "it is open to a respondent to advance any argument to sustain the judgment below, and he is not limited to appellants' points of law."

The mens rea issue is therefore properly before this Court.

V. Sections 32(1)(c) and 32(1.1) of the Act and Vagueness under Section 7 of the Charter

Since vagueness is a central issue in this case, it is useful to review the relevant principles and their application before dealing with the merits of the case.

A. Vagueness under the Canadian Charter of Rights and Freedoms

1. The Case Law of this Court

So far eight cases have given rise to discussions of vagueness issues under the Charter. A review of them will show that, while the theme of vagueness and the related notion of overbreadth have appeared in many decisions of this Court, giving rise to some questions as to the proper place of these concepts within Charter analysis, few statements have been made to substantiate the notion of vagueness, and its relationship with overbreadth.

Beetz J., in his opinion in R. v. Morgentaler, 1988 CanLII 90 (SCC), [1988] 1 S.C.R. 30, responded to the argument that s. 251(4)(c) of the Criminal Code, R.S.C. 1970, c. C‑34, was vague and offended s. 7 of the Charter. This section made abortions conditional upon the obtention of a doctor's certificate to the effect that the life or health of the woman was in danger. Beetz J. held that the standard of "likely danger to health" was not unduly vague. Since the law contemplated that the danger to health would be assessed by a medical practitioner exercising a medical judgment, some measure of flexibility was acceptable. "Flexibility and vagueness are not synonymous", wrote Beetz J. at p. 107.

In Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927, it was submitted that ss. 248 and 249 of the Quebec Consumer Protection Act, R.S.Q., c. P-40.1, were too vague to constitute a limit prescribed by law. Sections 248 and 249 forbid commercial advertising directed at persons under 13 years of age. Section 249 lists three factors to be taken into account when determining whether an advertisement was so directed, the last of which is the time and place of the advertisement. Then s. 249 enunciates that the mere fact that the advertisement was printed or broadcast in circumstances where it was intended for persons 13 and over or for all persons does not create a presumption that the advertisement is not directed at persons under 13. It was argued that s. 249 was confusing and left too much scope for discretion. The majority of the Court disagreed. It found that the text of the section could be given a sensible construction. On the issue of discretion, Dickson C.J., Lamer J. (as he then was) and Wilson J. wrote at p. 983:

Absolute precision in the law exists rarely, if at all. The question is whether the legislature has provided an intelligible standard according to which the judiciary must do its work. The task of interpreting how that standard applies in particular instances might always be characterized as having a discretionary element, because the standard can never specify all the instances in which it applies. On the other hand, where there is no intelligible standard and where the legislature has given a plenary discretion to do whatever seems best in a wide set of circumstances, there is no "limit prescribed by law".

Then came the Prostitution Reference, supra, where it was alleged that ss. 193 and 195.1(1)(c) of the Criminal Code, R.S.C. 1970, c. C-34, were impermissibly vague under s. 7 of the Charter. Lamer J. (as he then was) devoted a long passage in his reasons to the "void for vagueness" doctrine. Lamer J. ascribed two rationales for the invalidation of vague laws under s. 7 of the Charter at p. 1152, that is the need (1) to give citizens fair notice of the consequences of their conduct, so that they may avoid liability and benefit from a full answer and defence should they be tried and (2) to limit law enforcement discretion. He then reviewed the distinction between vagueness and overbreadth. At page 1155, Lamer J. pointed out that vagueness has been argued both under s. 7 and s. 1 of the Charter. He made some remarks on the issue: the vagueness doctrine does not require absolute certainty of laws, the interpretative role of the courts must not be overlooked and the possibility of varying interpretations is not fatal (at pp. 1156-57). He then proceeded to consider the impugned sections of the Code and found them not in violation of s. 7 of the Charter on account of vagueness. For the majority, Dickson C.J. endorsed Lamer J.'s analysis. While in dissent, Wilson J. agreed with the majority on this point.

Lamer J. also stated that even if the section was not unconstitutionally vague, it could nevertheless be found overly broad under s. 1 analysis. The majority did not consider this to be the case, but Wilson J. found the provisions too broad to meet the "minimal impairment" test. The Prostitution Reference established the doctrine of vagueness as one of the fundamental principles of justice under s. 7 of the Charter, and also differentiated vagueness and overbreadth.

This distinction came to the fore in R. v. Keegstra, 1990 CanLII 24 (SCC), [1990] 3 S.C.R. 697. This time the question was considered under the "minimal impairment" branch of the Oakes test (developed in R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103). It had been argued that s. 319(2) of the Criminal Code, R.S.C., 1985, c. C-46, was too broad in its ambit ("wilfully promot[ing] hatred against any identifiable group"), and could include expression that bears no relationship with Parliament's objective in enacting the statute. Furthermore, the vagueness of s. 319(2) was also raised. For the majority, Dickson C.J. blended both concepts. After a careful analysis of the wording of s. 319(2) and of the defences open to the accused in s. 319(3), Dickson C.J. concluded that the subsection did pass the "minimal impairment" test. McLachlin J. in dissent focused more precisely on the overbreadth of the section in her reasons, even though she relied on the same elements as Dickson C.J. Keegstra does not bring anything new to the principles that had already been developed, and the majority opinion does not distinguish between vagueness and overbreadth quite as clearly as in the Prostitution Reference.

In Canada (Human Rights Commission) v. Taylor, 1990 CanLII 26 (SCC), [1990] 3 S.C.R. 892, a companion case to Keegstra, McLachlin J. in dissent commented on the proper place of vagueness within Charter analysis. She wrote at p. 956:

[T]he difficulty in ascribing a constant and universal meaning to the terms used [in the impugned section] is a factor to be taken into account in assessing whether the law is "demonstrably justified in a free and democratic society". But I would be reluctant to circumvent the entire balancing analysis of the s. 1 test by finding that the words used were so vague as not to constitute a "limit prescribed by law", unless the provision could truly be described as failing to offer an intelligible standard.

McLachlin J. seems to envision a relatively stringent standard for vagueness, if it is to cut short the s. 1 analysis through a finding that the disposition is not law within the meaning of "prescribed by law".

In Committee for the Commonwealth of Canada v. Canada, 1991 CanLII 119 (SCC), [1991] 1 S.C.R. 139, L'Heureux-Dubé J. discussed the vagueness and overbreadth of s. 7 of the Government Airport Concession Operations Regulations, SOR/79-373. She was of the opinion that a statutory disposition, if too vague, would not constitute a limit "prescribed by law", while if overbroad, may not pass the Oakes test (at p. 208). In her discussion of vagueness, she linked this concept with the rule of law, and restated the dual concerns of fair notice and limitation of enforcement discretion that had been formulated by Lamer J. in the Prostitution Reference. As for overbreadth, L'Heureux-Dubé J. considered that this doctrine was more or less subsumed within the "minimal impairment" branch of the Oakes test.

Further in Osborne v. Canada (Treasury Board), 1991 CanLII 60 (SCC), [1991] 2 S.C.R. 69, the vagueness and overbreadth of s. 33 of the Public Service Employment Act, R.S.C., 1985, c. P-33, were at issue under the s. 1 analysis. For the majority, Sopinka J. adopted the statements made by McLachlin J. in Taylor and added at pp. 94-95:

Vagueness can have constitutional significance in at least two ways in a s. 1 analysis. A law may be so uncertain as to be incapable of being interpreted so as to constitute any restraint on governmental power. The uncertainty may arise either from the generality of the discretion conferred on the donee of the power or from the use of language that is so obscure as to be incapable of interpretation with any degree of precision using the ordinary tools. In these circumstances, there is no "limit prescribed by law" and no s. 1 analysis is necessary as the threshold requirement for its application is not met. The second way in which vagueness can play a constitutional role is in the analysis of s. 1. A law which passes the threshold test may, nevertheless, by reason of its imprecision, not qualify as a reasonable limit. Generality and imprecision of language may fail to confine the invasion of a Charter right within reasonable limits. In this sense vagueness is an aspect of overbreadth.

In its recent decision in R. v. Butler, 1992 CanLII 124 (SCC), [1992] 1 S.C.R. 452, this Court followed its case law and found that the words "undue exploitation of sex" in s. 163(8) of the Criminal Code, R.S.C., 1985, c. C-46, constituted a limit prescribed by law within the meaning of s. 1 of the Charter and, as interpreted by the courts satisfied the minimum impairment branch of the s. 1 test.

The foregoing may be summarized by way of the following propositions:

1. Vagueness can be raised under s. 7 of the Charter, since it is a principle of fundamental justice that laws may not be too vague. It can also be raised under s. 1 of the Charter in limine, on the basis that an enactment is so vague as not to satisfy the requirement that a limitation on Charter rights be "prescribed by law". Furthermore, vagueness is also relevant to the "minimal impairment" stage of the Oakes test (Morgentaler, Irwin Toy and the Prostitution Reference).

2. The "doctrine of vagueness" is founded on the rule of law, particularly on the principles of fair notice to citizens and limitation of enforcement discretion (Prostitution Reference and Committee for the Commonwealth of Canada).

3. Factors to be considered in determining whether a law is too vague include (a) the need for flexibility and the interpretative role of the courts, (b) the impossibility of achieving absolute certainty, a standard of intelligibility being more appropriate and (c) the possibility that many varying judicial interpretations of a given disposition may exist and perhaps coexist (Morgentaler, Irwin Toy, Prostitution Reference, Taylor and Osborne).

4. Vagueness, when raised under s. 7 or under s. 1 in limine, involves similar considerations (Prostitution Reference, Committee for the Commonwealth of Canada). On the other hand, vagueness as it relates to the "minimal impairment" branch of s. 1 merges with the related concept of overbreadth (Committee for the Commonwealth of Canada and Osborne).

5. The Court will be reluctant to find a disposition so vague as not to qualify as "law" under s. 1 in limine, and will rather consider the scope of the disposition under the "minimal impairment" test (Taylor and Osborne).

In order to give a more complete picture of issues of vagueness under the Charter, I will examine in turn the proper place of the doctrine of vagueness in Charter analysis and its content.

2. The Proper Place of the Doctrine of Vagueness in Charter Adjudication

Vagueness is often mingled and confused with overbreadth, possibly because of the influence of American authorities. From a review of American law, it will appear that overbreadth is not an autonomous notion in Canadian law, and that, contrary to the position of U.S. constitutional law, vagueness should have a constant meaning in Canadian law.

Overbreadth in American law is tied to the First Amendment. It is grounds to obtain what is termed "facial invalidation" of a statute, as opposed to a declaration that the statute is unconstitutional in the case of the particular plaintiff, which is the usual remedy. Village of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489 (1982), indicates how overbreadth interacts with vagueness in First Amendment cases. The court wrote at pp. 494-95:

In a facial challenge to the overbreadth and vagueness of a law, a court's first task is to determine whether the enactment reaches a substantial amount of constitutionally protected conduct. If it does not, then the overbreadth challenge must fail. The court should then examine the facial vagueness challenge and, assuming the enactment implicates no constitutionally protected conduct, should uphold the challenge only if the enactment is impermissibly vague in all of its applications.

Overbreadth ties in to the taxonomy of protected and unprotected conduct and expression developed by American courts under the First Amendment. Some conduct or expression receives First Amendment protection and some does not, and to the extent that a statute substantially touches upon protected conduct and cannot be severed or read down, it will be declared void (see L. H. Tribe, American Constitutional Law (2nd ed. 1988), at p. 1022).

This distinction between protected and unprotected conduct or expression is typical of American law, since the American Constitution does not contain a general balancing clause similar to s. 1 of the Charter. Balancing must be done within the First Amendment itself. In this respect, it can be seen that the doctrine of overbreadth in American law involves an element of balancing, since the aims and scope of the statute must be compared with the range of protection of the First Amendment. C. Rogerson, "The Judicial Search for Appropriate Remedies Under the Charter: The Examples of Overbreadth and Vagueness", in R. J. Sharpe, Charter Litigation (1987), at pp. 261-62, traces this element of balancing to Broadrick v. Oklahoma, 413 U.S. 601 (1973).

This Court has repeatedly emphasized the numerous differences which exist between the Charter and the American Constitution. In particular, in the interpretation of s. 2 of the Charter, this Court has taken a route completely different from that of U.S. courts. In cases starting with Irwin Toy up to Butler, including the Prostitution Reference and Keegstra, this Court has given a wide ambit to the freedoms guaranteed by s. 2 of the Charter, on the basis that balancing between the objectives of the State and the violation of a right or freedom should occur at the s. 1 stage. Other sections of the Charter, such as ss. 7 and 8, do however incorporate some element of balancing, as a limitation within the definition of the protected right, with respect to other notions such as principles of fundamental justice or reasonableness.

A notion tied to balancing such as overbreadth finds its proper place in sections of the Charter which involve a balancing process. Consequently, I cannot but agree with the opinion expressed by L'Heureux-Dubé J. in Committee for the Commonwealth of Canada that overbreadth is subsumed under the "minimal impairment branch" of the Oakes test, under s. 1 of the Charter. This is also in accordance with the trend evidenced in Osborne and Butler. Furthermore, in determining whether s. 12 of the Charter has been infringed, for instance, a court, if it finds the punishment not grossly disproportionate for the accused, will typically examine reasonable hypotheses and assess whether the punishment is grossly disproportionate in these situations (R. v. Smith, 1987 CanLII 64 (SCC), [1987] 1 S.C.R. 1045, R. v. Goltz, 1991 CanLII 51 (SCC), [1991] 3 S.C.R. 485). This inquiry also resembles the sort of balancing process associated with the notion of overbreadth.

In all these cases, however, overbreadth remains no more than an analytical tool. The alleged overbreadth is always related to some limitation under the Charter. It is always established by comparing the ambit of the provision touching upon a protected right with such concepts as the objectives of the State, the principles of fundamental justice, the proportionality of punishment or the reasonableness of searches and seizures, to name a few. There is no such thing as overbreadth in the abstract. Overbreadth has no autonomous value under the Charter. As will be seen below, overbreadth is not at the heart of this case, although it has been invoked in argument.

The relationship between vagueness and "overbreadth" was well expounded by the Ontario Court of Appeal in this oft-quoted passage from R. v. Zundel (1987), 1987 CanLII 121 (ON CA), 58 O.R. (2d) 129, at pp. 157-58:

Vagueness and overbreadth are two concepts. They can be applied separately, or they may be closely interrelated. The intended effect of a statute may be perfectly clear and thus not vague, and yet its application may be overly broad. Alternatively, as an example of the two concepts being closely interrelated, the wording of a statute may be so vague that its effect is considered to be overbroad.

I agree. A vague law may also constitute an excessive impairment of Charter rights under the Oakes test. This Court recognized this, when it mentioned the two aspects of vagueness under s. 1 of the Charter, in Osborne and Butler.

For the sake of clarity, I would prefer to reserve the term "vagueness" for the most serious degree of vagueness, where a law is so vague as not to constitute a "limit prescribed by law" under s. 1 in limine. The other aspect of vagueness, being an instance of overbreadth, should be considered as such.

Moreover, in American constitutional law, "vagueness" has been given various contents, depending on whether it is considered under the First, Fifth (and Fourteenth) or even Eighth Amendment (for statutory conditions warranting the imposition of the death penalty). The scope of the inquiry and the reference group will change, as the United States Supreme Court stated in Maynard v. Cartwright, 486 U.S. 356 (1988), at p. 361.

Under the Charter, however, given the statements of Lamer J. in the Prostitution Reference, at p. 1155, I would consider that the "doctrine of vagueness" is a single concept, whether invoked as a principle of fundamental justice under s. 7 of the Charter or as part of s. 1 of the Charter in limine. Indeed from a practical point of view this makes little difference in the analysis, since a consideration of s. 1 in limine would follow immediately the determination of whether s. 7 has been violated. No intermediate step is lost. From a theoretical perspective, the justifications invoked for the doctrine of vagueness under both s. 7 and s. 1 are similar. A reading of the aforementioned cases shows that the rationales of fair notice to the citizen and limitation of enforcement discretion are put forward in every discussion of vagueness, irrespective of where it occurs in the Charter analysis. I see no ground for distinguishing them.

Vagueness may be raised under the substantive sections of the Charter whenever these sections comprise some internal limitation. For example, under s. 7, it may be that the limitation on life, liberty and security of the person would not otherwise be objectionable, but for the vagueness of the impugned law. The doctrine of vagueness would then rank among the principles of fundamental justice. Outside of these cases, the proper place of a vagueness argument is under s. 1 in limine.

I would therefore conclude that:

1. What is referred to as "overbreadth", whether it stems from the vagueness of a law or from another source, remains no more than an analytical tool to establish a violation of a Charter right. Overbreadth has no independent existence. References to a "doctrine of overbreadth" are superfluous.

2. The "doctrine of vagueness", the content of which will be developed shortly, is a principle of fundamental justice under s. 7 and it is also part of s. 1 in limine ("prescribed by law").

3. The Content of the "Doctrine of Vagueness"

As was said by this Court in Osborne and Butler, the threshold for finding a law vague is relatively high. So far discussion of the content of the notion has evolved around intelligibility.

The two rationales of fair notice to the citizen and limitation of enforcement discretion have been adopted as the theoretical foundations of the doctrine of vagueness, here (Prostitution Reference and Committee for the Commonwealth of Canada) as well as in the United States (see Grayned v. City of Rockford, 408 U.S. 104 (1972), at pp. 108-9) and in Europe, as will be seen later. These two rationales have been broadly linked with the corpus of principles of government known as the "rule of law", which lies at the core of our political and constitutional tradition.

(a) Fair Notice to the Citizen

Fair notice to the citizen, as a guide to conduct and a contributing element to a full answer and defence, comprises two aspects.

First of all, there is the more formal aspect of notice, that is acquaintance with the actual text of a statute. In the criminal context, this concern has more or less been set aside by the common law maxim "Ignorance of the law is no excuse", embodied in s. 19 of the Criminal Code (see R. v. MacDougall, 1982 CanLII 212 (SCC), [1982] 2 S.C.R. 605). In the civil context, the maxim does not apply with equal force (see J.‑L. Baudouin, Les obligations (3rd ed. 1989), at p. 122, and Chitty on Contracts (25th ed. 1983), at paras. 314 and 353). Some authors have expressed the opinion that this maxim contradicts the rule of law, and should be revised in the light of the growing quantity and complexity of penal legislation (see E. Colvin, "Criminal Law and The Rule of Law", in P. Fitzgerald, ed., Crime, Justice & Codification (1986), 125, at p. 151, and J. C. Jeffries, Jr., "Legality, Vagueness, and the Construction of Penal Statutes" (1985), 71 Va. L. Rev. 189, at p. 209). Since this argument was not raised in this case, I will refrain from ruling on this issue. In any event, given that, as this Court has already recognized, case law applying and interpreting a particular section is relevant in determining whether the section is vague, formal notice is not a central concern in a vagueness analysis.

As Lamer J. pointed out in Re B.C. Motor Vehicle Act, supra, principles of fundamental justice, such as the doctrine of vagueness, must have a substantive as well as procedural content. Indeed the idea of giving fair notice to citizens would be rather empty if the mere fact of bringing the text of the law to their attention was enough, especially when knowledge is presumed by law. There is also a substantive aspect to fair notice, which could be described as a notice, an understanding that some conduct comes under the law. Jeffries, supra, calls this the "core concept of notice" (p. 211).

Let me take homicide as an example. The actual provisions of the Criminal Code dealing with homicide are numerous (comprising the core of ss. 222-240 and other related sections). When one completes the picture of the Code with case law, both substantive and constitutional, the result is a fairly intricate body of rules. Notwithstanding formal notice, it can hardly be expected of the average citizen that he know the law of homicide in detail. Yet no one would seriously argue that there is no substantive fair notice here, or that the law of homicide is vague. It can readily be seen why this is so. First of all, everyone (or sadly, should I say, almost everyone) has an inherent knowledge that taking the life of another human being is wrong. There is a deeply-rooted perception that homicide cannot be tolerated, whether one comes to this perception from a moral, religious or sociological stance. Therefore it is expected that homicide will be punished by the State. Secondly, homicide is indeed punished by the State, and homicide trials and sentences receive a great deal of publicity.

I used homicide as an example, because it lies so at the core of our criminal law and our shared values that substantive notice is easy to demonstrate. Similar demonstrations could be made, at greater length, for other legal provisions. The substantive aspect of fair notice is therefore a subjective understanding that the law touches upon some conduct, based on the substratum of values underlying the legal enactment and on the role that the legal enactment plays in the life of the society.

I do not wish to suggest that the State can only intervene through law when some non-legal basis for intervention exists. Many enactments are relatively narrow in scope and echo little of society at large; this is the case with many regulatory enactments. The weakness or the absence of substantive notice before the enactment can be compensated by bringing to the attention of the public the actual terms of the law, so that substantive notice will be achieved. Merit point and driving license revocation schemes are prime examples of this; through publicity and advertisement these schemes have been "digested" by society. A certain connection between the formal and substantive aspects of fair notice can be seen here.

Fair notice may not have been given when enactments are in somewhat general terms, in a way that does not readily permit citizens to be aware of their substance, when they do not relate to any element of the substratum of values held by society. It is no coincidence that these enactments are often found vague. For instance, the vagrancy ordinance invalidated by the United States Supreme Court in Papachristou v. City of Jacksonville, 405 U.S. 156 (1972), or the compulsory identification statute struck down in Kolender v. Lawson, 461 U.S. 352 (1983), fall in this group.

Hence, aside from a formal aspect which is in our current system often presumed, fair notice to the citizen comprises a substantive aspect, that is an understanding that certain conduct is the subject of legal restrictions.

(b) Limitation of Law Enforcement Discretion

Lamer J. in the Prostitution Reference used the phrase "standardless sweep", first coined by the United States Supreme Court in Smith v. Goguen, 415 U.S. 566 (1974), at p. 575, to describe the limitation of enforcement discretion rationale for the doctrine of vagueness. It has become the prime concern in American constitutional law (Kolender, at pp. 357-58). Indeed today it has become paramount, given the considerable expansion in the discretionary powers of enforcement agencies that has followed the creation of the modern welfare State.

A law must not be so devoid of precision in its content that a conviction will automatically flow from the decision to prosecute. Such is the crux of the concern for limitation of enforcement discretion. When the power to decide whether a charge will lead to conviction or acquittal, normally the preserve of the judiciary, becomes fused with the power to prosecute because of the wording of the law, then a law will be unconstitutionally vague.

For instance, the wording of the vagrancy ordinance invalidated by the United States Supreme Court in Papachristou and quoted at length in the Prostitution Reference, at pp. 1152-53, was so general and so lacked precision in its content that a conviction would ensue every time the law enforcer decided to charge someone with the offence of vagrancy. The words of the ordinance had no substance to them, and they indicated no particular legislative purpose. They left the accused completely in the dark, with no possible way of defending himself before the court.

(c) European Court of Human Rights Case Law

I would also note that the European Court of Human Rights (hereinafter "ECHR") has adopted the same approach to issues of vagueness, in the course of its treatment of words such as "prescribed by law", found in many limitation clauses of the European Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 222 (hereinafter the "Convention"), such as Articles 8(2), 9(2), 10(2), 11(2) or Article 2(3) of Protocol No. 4 to the Convention, Europ. T.S. No. 46. The ECHR gave this phrase a substantive content, which went beyond a mere inquiry as to whether a law existed or not.

The ECHR developed its conception of "prescribed by law" in the course of two famous cases, the Sunday Times case, judgment of 26 April 1979, Series A No. 30, and the Malone case, judgment of 2 August 1984, Series A No. 82. In the former, the ECHR drew attention to the two aspects of fair notice, namely formal notice ("accessibility") and substantive notice ("foreseeability"). It wrote at p. 31:

In the Court's opinion, the following are two of the requirements that flow from the expression "prescribed by law". Firstly, the law must be adequately accessible: the citizen must be able to have an indication that is adequate in the circumstances of the legal rules applicable to a given case. Secondly, a norm cannot be regarded as a "law" unless it is formulated with sufficient precision to enable the citizen to regulate his conduct: he must be able -- if need be with appropriate advice -- to foresee, to a degree that is reasonable in the circumstances, the consequences which a given action may entail. Those consequences need not be foreseeable with absolute certainty: experience shows this to be unattainable. Again, whilst certainty is highly desirable, it may bring in its train excessive rigidity and the law must be able to keep pace with changing circumstances. Accordingly, many laws are inevitably couched in terms which, to a greater or lesser extent, are vague and whose interpretation and application are questions of practice.

In the latter, the ECHR added the limitation of enforcement discretion to the range of interests underpinning its interpretation of "prescribed by law" at p. 32:

The phrase thus implies . . . that there must be a measure of legal protection in domestic law against arbitrary interferences by public authorities with the rights safeguarded by paragraph 1 [of article 8 of the Convention].

(See also the Kruslin case, judgment of 24 April 1990, Series A No. 176-A, at pp. 24‑25, and the Huvig case, judgment of 24 April 1990, Series A No. 176-B, at p. 56.)

In my opinion, the case law of the ECHR is a very valuable guide on this issue, and it will be relied on further below.

(d) The Scope of Precision

This leads me to synthetize these remarks about vagueness. The substantive notice and limitation of enforcement discretion rationales point in the same direction: an unintelligible provision gives insufficient guidance for legal debate and is therefore unconstitutionally vague.

Legal provisions by stating certain propositions outline certain permissible and impermissible areas, and they also provide some guidance to ascertain the boundaries of these areas. In his survey article "La teneur indécise du droit" (1991), 107 Rev. dr. publ. 1199, P. Amselek rightly underlines the etymological and metaphorical relationship between law and geometry and writes at pp. 1200-1201:

[translation] Legal rules are mental tools . . . authoritatively introduced, given effect, by public authorities placed at the head of human communities to govern them: such rules are thought content with a specific purpose, to be used as a tool to guide conduct, thought content which determines the boundaries of possible action depending on the circumstances ‑‑ for the Romans, these boundaries were the meaning of the very concept of jus in its earliest sense and are also reflected in our concept of "law", implying the very idea of possibility, of latitude. These boundaries impose limits on the will of those to whom they apply, serving as a support, a yardstick enabling them to remain within the area of right conduct, of rectitude, within the parameters of conduct which the concept lays down and which it then gives effect to, setting the process in motion.

These rules, as Amselek later points out, are characterized by their unresolved nature, inasmuch as they are neither objective nor complete.

Legal rules only provide a framework, a guide as to how one may behave, but certainty is only reached in instant cases, where law is actualized by a competent authority. In the meanwhile, conduct is guided by approximation. The process of approximation sometimes results in quite a narrow set of options, sometimes in a broader one. Legal dispositions therefore delineate a risk zone, and cannot hope to do more, unless they are directed at individual instances.

By setting out the boundaries of permissible and non-permissible conduct, these norms give rise to legal debate. They bear substance, and they allow for a discussion as to their actualization. They therefore limit enforcement discretion by introducing boundaries, and they also sufficiently delineate an area of risk to allow for substantive notice to citizens.

Indeed no higher requirement as to certainty can be imposed on law in our modern State. Semantic arguments, based on a perception of language as an unequivocal medium, are unrealistic. Language is not the exact tool some may think it is. It cannot be argued that an enactment can and must provide enough guidance to predict the legal consequences of any given course of conduct in advance. All it can do is enunciate some boundaries, which create an area of risk. But it is inherent to our legal system that some conduct will fall along the boundaries of the area of risk; no definite prediction can then be made. Guidance, not direction, of conduct is a more realistic objective. The ECHR has repeatedly warned against a quest for certainty and adopted this "area of risk" approach in Sunday Times, supra, and especially the case of Silver and others, judgment of 25 March 1983, Series A No. 61, at pp. 33-34, and Malone, supra, at pp. 32-33.

A vague provision does not provide an adequate basis for legal debate, that is for reaching a conclusion as to its meaning by reasoned analysis applying legal criteria. It does not sufficiently delineate any area of risk, and thus can provide neither fair notice to the citizen nor a limitation of enforcement discretion. Such a provision is not intelligible, to use the terminology of previous decisions of this Court, and therefore it fails to give sufficient indications that could fuel a legal debate. It offers no grasp to the judiciary. This is an exacting standard, going beyond semantics. The term "legal debate" is used here not to express a new standard or one departing from that previously outlined by this Court. It is rather intended to reflect and encompass the same standard and criteria of fair notice and limitation of enforcement discretion viewed in the fuller context of an analysis of the quality and limits of human knowledge and understanding in the operation of the law.

(e) Vagueness and the Rule of Law

The criterion of absence of legal debate relates well to the rule of law principles that form the backbone of our polity. Here one must see the rule of law in the contemporary context. Continental European studies on the "État de droit" or "Rechtsstaat" are relevant (see L. C. Blaau, "The Rechtsstaat Idea Compared with the Rule of Law as a Paradigm for Protecting Rights" (1990), 107 S. Afr. L.J. 76, at pp. 88-92, for an exposition of the historical differences between these concepts).

J.-P. Henry, "Vers la fin de l'État de droit?" (1977), 93 Rev. dr. publ. 1207, gives the following definition of the "État de droit" at p. 1208:

[translation] In theoretical terms, the État de droit is a system of organization in which all social and political relations are subject to the law. This means that relations between individuals and authority, as well as relations between individuals themselves, are part of a legal interchange involving rights and obligations.

See also J. Chevallier, "L'État de droit" (1988), 104 Rev. dr. publ. 313, at pp. 330-31, and R. Carré de Malberg, Contribution à la théorie générale de l'État (1920), vol. 1, at pp. 488‑90. At the core of the "État de droit", as under the rule of law, lies the proposition that the relationship of the State to the individuals is regulated by law.

One must move away from the non-interventionist attitude that surrounded the development of the doctrine of the rule of law to a more global conception of the State as an entity bound by and acting through law. The modern State intervenes in almost every field of human endeavour, and it plays a role that goes far beyond collecting taxes and policing. The State has entered fields where the positions are not so clear-cut; in the realm of social or economic policy, interests diverge, and the State does not seek to enforce a definite and limited social interest in public order, for instance, against an individual. Often the State attempts to realize a series of social objectives, some of which must be balanced against one another, and which sometimes conflict with the interests of individuals. The modern State, while still acting as an enforcer, assumes more and more of an arbitration role.

This arbitration must be done according to law, but often it reaches such a level of complexity that the corresponding enactment will be framed in relatively general terms. In my opinion the generality of these terms may entail a greater role for the judiciary, but unlike some authors (see F. Neumann, The Rule of Law (1986), at pp. 238-39), I fail to see a difference in kind between general provisions where the judiciary would assume part of the legislative role and "mechanical" provisions where the judiciary would simply apply the law. The judiciary always has a mediating role in the actualization of law, although the extent of this role may vary.

Indeed, as the ECHR has recognized in Sunday Times, supra, and particularly in the Barthold case, judgment of 25 March 1985, Series A No. 90, at p. 22, and in the case of Müller and others, judgment of 24 May 1988, Series A No. 133, at p. 20, laws that are framed in general terms may be better suited to the achievement of their objectives, inasmuch as in fields governed by public policy circumstances may vary widely in time and from one case to the other. A very detailed enactment would not provide the required flexibility, and it might furthermore obscure its purposes behind a veil of detailed provisions. The modern State intervenes today in fields where some generality in the enactments is inevitable. The substance of these enactments remains nonetheless intelligible. One must be wary of using the doctrine of vagueness to prevent or impede State action in furtherance of valid social objectives, by requiring the law to achieve a degree of precision to which the subject-matter does not lend itself. A delicate balance must be maintained between societal interests and individual rights. A measure of generality also sometimes allows for greater respect for fundamental rights, since circumstances that would not justify the invalidation of a more precise enactment may be accommodated through the application of a more general one.

What becomes more problematic is not so much general terms conferring broad discretion, but terms failing to give direction as to how to exercise this discretion, so that this exercise may be controlled. Once more, an unpermissibly vague law will not provide a sufficient basis for legal debate; it will not give a sufficient indication as to how decisions must be reached, such as factors to be considered or determinative elements. In giving unfettered discretion, it will deprive the judiciary of means of controlling the exercise of this discretion. The need to provide guidelines for the exercise of discretion was at the centre of the ECHR reasons in Malone, supra, at pp. 32-33, and the Leander case, judgment of 26 March 1987, Series A No. 116, at p. 23.

Finally, I also wish to point out that the standard I have outlined applies to all enactments, irrespective of whether they are civil, criminal, administrative or other. The citizen is entitled to have the State abide by constitutional standards of precision whenever it enacts legal dispositions. In the criminal field, it may be thought that the terms of the legal debate should be outlined with special care by the State. In my opinion, however, once the minimal general standard has been met, any further arguments as to the precision of the enactments should be considered at the "minimal impairment" stage of s. 1 analysis.

The doctrine of vagueness can therefore be summed up in this proposition: a law will be found unconstitutionally vague if it so lacks in precision as not to give sufficient guidance for legal debate. This statement of the doctrine best conforms to the dictates of the rule of law in the modern State, and it reflects the prevailing argumentative, adversarial framework for the administration of justice.

B. The Validity of Section 32(1)(c) of the Act

The offence created by s. 32(1)(c) comprises two material elements:

1. An agreement entered into by the accused ("Every one who conspires, combines, agrees or arranges with another person"); and

2. An undue prevention or lessening of competition flowing from this agreement ("to prevent, or lessen, unduly, competition in the production, manufacture, purchase, barter, sale, storage, rental, transportation or supply of a product, or in the price of insurance upon persons or property").

There is furthermore a mental element to this offence, which I will discuss in Part VI of these reasons.

The first element has given rise to some debate throughout the history of competition legislation in Canada, but it is not the prime concern of this appeal. The bulk of the argument before us has been on the second element of s. 32(1)(c), more precisely on the word "unduly". Only the interveners Association québécoise des pharmaciens propriétaires et al. (hereinafter "AQPP") presented submissions on other points, and I will deal with them briefly before concentrating on the word "unduly", that is to say, on the test for differentiating between agreements which fall under s. 32(1)(c) of the Act and others which do not.

First of all, the AQPP has argued that the range of agreements covered by s. 32(1)(c) is too wide. This submission bears more on "overbreadth" than vagueness, and it shows the dangers of too great a reliance on this concept. In order to claim that s. 32(1)(c) covers too many types of agreements, some idea as to its proper scope must be explicitly or implicitly advanced. Section 32(1)(c) must be too broad in the light of some Charter right, if one is to find a Charter violation. Setting aside the vagueness issue, the AQPP does not argue that s. 32(1)(c) imposes a grossly disproportionate punishment, jeopardizes the impartiality of the tribunal, or interferes with some other Charter right. The AQPP might find that the scope of the section is too broad with respect to the objectives of the State in fostering free competition. Absent a violation of the Charter, though, the mere fact that the State may have reached for more than its objectives might have warranted is no ground for constitutional redress. In effect, the AQPP argument rests on an implicit assumption that some agreements or some persons are entitled to escape the application of competition law, presumably because of their smallness or their innocuity. Our constitution knows of no right to be shielded from economic regulation on such grounds. Claims of overbreadth should not be used to masquerade an absence of constitutional foundation. In the instant case any excess in the scope of s. 32(1)(c) would only be the result of the alleged vagueness of the word "unduly", and claims of overbreadth will succeed or fail accordingly.

Furthermore, the AQPP has claimed that the Act, in giving enforcement authorities, under certain circumstances, a discretion between penal and civil recourses, leaves them with too much discretion. The source of this allegedly excessive discretion lies in the structure of the Act and not in s. 32(1)(c) of the Act itself. This claim lies beyond the lis of this case, but I will nevertheless address it. The AQPP argues that the Act, because of the extent of overlap between its central provisions dealing with conspiracies, abuses of dominant position and mergers, confers on the Director of Investigation and Research an excessive discretion to choose between so-called "civil" recourses before the Competition Tribunal and criminal recourses. The AQPP further adds that s. 30(2) of the Act (now s. 34(2)) gives the Director a choice between a criminal prosecution and a prohibition order, when seeking relief before the criminal courts.

In the first place, what the AQPP brands as an arbitrary power given to the Director does not correspond to the kind of excessive enforcement discretion leading to concerns about the vagueness of the law. It is not submitted that the law, because of its imprecision, essentially gives the Director a power to convict from the moment a prosecution is brought against a person under the Act. The choice between criminal and civil/administrative remedies is and remains the sole concern of the AQPP, and this concern does not relate to the doctrine of vagueness. The thrust of the AQPP is more akin to double jeopardy. In this respect, it must be said that, except for s. 32(1)(c), the options open to the Director do not qualify for the application of s. 11(h), following R. v. Wigglesworth, 1987 CanLII 41 (SCC), [1987] 2 S.C.R. 541, and R. v. Shubley, 1990 CanLII 149 (SCC), [1990] 1 S.C.R. 3. Moreover, the Charter does not prevent Parliament from creating offences that may overlap. The guarantees against double jeopardy found in s. 11(h) and perhaps also s. 7 of the Charter apply only to proceedings, not to legal enactments. Even then, it is apparent that Parliament has added ss. 32.01, 51(7) and 70 to the Act (now ss. 45.1, 79(7) and 98) precisely in order to prevent multiple proceedings against a person on the same or a similar factual basis.

The additional arguments of the AQPP therefore fail, and I will now consider the alleged vagueness of s. 32(1)(c) of the Act.

1. "Unduly" in the Decisions of this Court

This Court has already interpreted the word "unduly" on a number of occasions, namely in Weidman v. Shragge (1912), 1912 CanLII 58 (SCC), 46 S.C.R. 1; Stinson-Reeb Builders Supply Co. v. The King, 1929 CanLII 83 (SCC), [1929] S.C.R. 276; Container Materials Ltd. v. The King, 1942 CanLII 1 (SCC), [1942] S.C.R. 147; Howard Smith Paper Mills Ltd. v. The Queen, supra; Aetna Insurance Co. v. The Queen, 1977 CanLII 174 (SCC), [1978] 1 S.C.R. 731, and Atlantic Sugar Refineries Co. v. Attorney General of Canada, supra. I will not proceed to a lengthy survey of these decisions (this has been done in Aetna Insurance).

The meaning of "unduly" has usually been described by reference to various synonyms. In Weidman, Anglin J. adopted the string of synonyms given in R. v. Elliott (1905), 1905 CanLII 117 (ON CA), 9 C.C.C. 505 (Ont. C.A.), at p. 520 ("improper, inordinate, excessive or oppressive"). This string has been mentioned in all subsequent judgments of this Court on the issue. In Howard Smith, Cartwright J. sought to restate the law and introduced what was branded as the "virtual elimination of competition" criterion. Cartwright J. was in minority in that case, and furthermore a close reading of his reasons reveals that he might not have intended to depart from the existing law (see the dissenting reasons of Laskin J.A. (as he then was) in R. v. J. J. Beamish Construction Co. (1967), 1967 CanLII 239 (ON CA), 65 D.L.R. (2d) 260 (Ont. C.A.), at p. 285). Nevertheless, given that a controversy had arisen, Parliament saw fit to add s. 32(1.1) to the Act, thus dispelling any doubt as to the state of the law. Anglin J.'s reasons in Weidman remain authoritative. Instead of adding to this string of synonyms, I will adopt the reasons of Clarke C.J.N.S. in the instant case at p. 157:

While the word unduly is not defined by statute and defies precise measurement, it is a word of common usage which denotes to all of us in one way or another a sense of seriousness. Something affected unduly is not affected to a minimal degree but to a significant degree.

Clarke C.J.N.S. has touched upon the very nature of "unduly": while this word does not have a precise technical meaning, it does have a common meaning, and it expresses a notion of seriousness or significance.

Nothing more will be gained by a semantic inquiry into the meaning of "unduly". Such an inquiry is a beginning, however, not an end. The appellants and the AQPP were content with pointing out the generality of "unduly" and arguing that any other consideration pertaining to s. 32(1)(c) is a question of fact, following Container Materials. According to the appellants, since the determination of whether the restriction on competition was undue is a question of fact, not subject to appellate review, no conclusion can be drawn from the case law. This argument rests on a mistaken perception of the distinction between questions of fact and questions of law.

In the context of s. 32(1)(c), the process followed and the criteria used to arrive at a determination of "undueness" are questions of law and as such are reviewable by an appellate court. The application of this process and these criteria, that is the full inquiry, often involving complicated economic issues, into whether the impugned agreement was an undue restriction on competition, remains a question of fact. The general rule that appellate courts should be reluctant to venture into a re‑examination of the factual conclusions of the trial judge applies with special force in a complex matter such as here.

The legal content of s. 32(1)(c), however, is not exhausted by a search for the meaning of unduly. Section 32(1)(c) must not be taken in a vacuum. Its interpretation is conditioned, first of all, by the purposes of the Act. Furthermore, its content is enriched by the rest of the section in which it is found and by the mode of inquiry adopted by courts as they have ruled under it. These are matters of law, pertaining to the determination of undueness under s. 32(1)(c), and as such most relevant. I will look at each in turn.

2. The Act, Section 32(1)(c) and Canadian Public Policy

At the outset, it must be noted that the Act is central to Canadian public policy in the economic sector, and that s. 32 is itself one of the pillars of the Act.

The history of competition legislation in Canada dates back to 1889, with the Act for the Prevention and Suppression of Combinations formed in restraint of Trade, S.C. 1889, c. 41 (hereinafter the "1889 Act"). In fact, the 1889 Act came into force before the American Sherman Act, c. 647, 26 Stat. 209 (1890), codified as amended at 15 U.S.C. §§ 1-7, generally seen as the primogenitor of competition law. The 1889 Act created in s. 1 the offence of "combining for the purpose of unlawfully hindering competition", in substance the direct ancestor to s. 32(1) of the Act.

The 1889 Act was perhaps the first major foray by Parliament in the realm of economic policy. As B. Dunlop, D. McQueen and M. Trebilcock, Canadian Competition Policy: A Legal and Economic Analysis (1987), point out at p. 42, the common law of restraint of trade was very much anchored within a private law framework, and was not concerned with broader interests in the proper functioning of the economy. The 1889 Act introduced these interests in Canadian law (see Container Materials, supra, at p. 152, per Duff C.J.).

The content of the 1889 Act was progressively enlarged, until it grew into the current Act. In General Motors of Canada Ltd. v. City National Leasing, 1989 CanLII 133 (SCC), [1989] 1 S.C.R. 641, Dickson C.J. wrote for the Court at p. 676:

From this overview of the Combines Investigation Act I have no difficulty in concluding that the Act as a whole embodies a complex scheme of economic regulation. The purpose of the Act is to eliminate activities that reduce competition in the market-place. The entire Act is geared to achieving this objective. The Act identifies and defines anti-competitive conduct. It establishes an investigatory mechanism for revealing prohibited activities and provides an extensive range of criminal and administrative redress against companies engaging in behaviour that tends to reduce competition.

The Act can thus be seen as a central and established feature of Canadian economic policy.

Section 32(1)(c) of the Act moreover is its oldest provision. Even today, it remains at the core of the criminal part of the Act. The prohibition of conspiracies in restraint of trade is the epitome of competition law, finding its place in every competition law, from §1 of the Sherman Act to Article 85 of the Treaty of Rome. Section 32(1)(c) of the Act is not just another regulatory provision. It definitely rests on a substratum of values, a finding which must be kept in mind in the course of the vagueness analysis.

This Court has made numerous remarks on the public policy interests underlying s. 32(1)(c) of the Act. These remarks, found in Weidman, Stinson-Reed, Container Materials, and Aetna Insurance, supra, are perhaps best summarized in this passage from the majority judgment in Howard Smith, supra, at p. 411:

The statute proceeds upon the footing that the preventing or lessening of competition is in itself an injury to the public. It is not concerned with public injury or public benefit from any other standpoint.

Considerations such as private gains by the parties to the agreement or counterbalancing efficiency gains by the public lie therefore outside of the inquiry under s. 32(1)(c). Competition is presumed by the Act to be in the public benefit. The only issue is whether the agreement impairs competition to the extent that it will attract liability.

The peculiar nature of the inquiry under s. 32(1)(c) of the Act becomes apparent when it is compared with § 1 of the Sherman Act. Since the inception of the Sherman Act, American antitrust law has developed the two paradigms of adjudication known as the "per se rule" and the "rule of reason". The former attaches consequences to precisely-defined acts, irrespective of surrounding circumstances, whereas the latter is more general and invites in-depth inquiry into the details of the operation. The distinction between the two is not airtight, as leading authors have shown (see 7 P. Areeda, Antitrust Law ¶ 1511 (1987 & Supp. 1990)).

Section 32(1)(c) of the Act lies somewhere on the continuum between a per se rule and a rule of reason. It does allow for discussion of the anti-competitive effects of the agreement, unlike a per se rule, which might dictate that all agreements that lessen competition attract liability. On the other hand, it does not permit a full-blown discussion of the economic advantages and disadvantages of the agreement, like a rule of reason would. Since "unduly" in s. 32(1)(c) leads to a discussion of the seriousness of the competitive effects, but not of all relevant economic matters, one may say that this section creates a partial rule of reason.

The public policy objectives of s. 32(1)(c) of the Act already offer a clear idea of what is meant by "unduly" lessening competition, and what kind of inquiry is mandated. In fact, s. 32(1)(c) embodies a general standard, much like art. 1053 of the Civil Code of Lower Canada or s. 219 of the Criminal Code, for that matter. It represents an intelligible principle, one that carries meaning and that has conceptual force. In all its generality, however, it cannot readily be applied to a factual situation to yield an answer. Few legal norms are so.

The accused may actually be favoured by having a clear statement of principle, inasmuch as it contains the "spirit" as well as the letter of the law. Even when considered without the rest of the Act and case law, I would be inclined to say that s. 32(1)(c) of the Act is sufficiently precise to meet the constitutional standard. It outlines an area of risk, agreements that lessen or prevent competition, and imposes some boundaries on enforcement discretion, inasmuch as courts will scrutinize the impact of the agreement on competition to see if it runs against our public policy of free competition.

3. The Content of the Inquiry under Section 32(1)(c) of the Act

In addition, s. 32(1)(c) is made even more precise when one considers the content of the inquiry it mandates.

Since the few cases that have been considered by this Court always involved agreements where the effects on competition were easily ascertainable, this Court has never had the opportunity to consider the process whereby the undueness of the restriction on competition is assessed. In the context of this Charter inquiry into the alleged vagueness of s. 32(1)(c) of the Act, a survey of the rest of the section, together with lower court decisions and doctrinal writings, will show that adjudication under s. 32(1)(c) follows a definite process that eliminates any vagueness that might remain.

First of all, there are two major elements to this inquiry, that is (1) the structure of the market and (2) the behaviour of the parties to the agreement. As a preliminary step, definition of the relevant market is required. Many decisions have explicitly adopted this approach (R. v. J. W. Mills & Son Ltd., 1968 CanLII 1279 (CA EXC), [1968] 2 Ex. C.R. 275, at p. 303; J. J. Beamish, supra, at pp. 271 and 273; R. v. Canadian Coat and Apron Supply Ltd., 1967 CanLII 1157 (CA EXC), [1967] 2 Ex. C.R. 53, at p. 68; R. v. Anthes Business Forms Ltd. (1975), 1975 CanLII 54 (ON CA), 26 C.C.C. (2d) 349 (Ont. C.A.), at pp. 375-76; R. v. Canadian General Electric Co. (1976), 1976 CanLII 756 (ON SC), 34 C.C.C. (2d) 489 (Ont. H.C.), at p. 500).

I will not venture into the intricacies of outlining the relevant market, other than to repeat that it comprises both geographical and product or service aspects, as was stated in J. W. Mills, at p. 303. Definition of the relevant market is a fairly circumscribed process, even though it may require considerable inquiry (see R. v. Metropolitan Toronto Pharmacists' Association (1984), 1984 CanLII 5940 (ON SC), 3 C.P.R. (3d) 233 (Ont. H.C.)).

The structure-behaviour framework of analysis remains merely a convenient way of approaching conspiracy problems, and it should not be seen as a rite of passage. Indeed to a certain extent the determination of whether an agreement unduly restricts competition involves an examination not only of market structure and firm behaviour separately, but also of the relationship between them, as Gibson J. remarked in J. W. Mills, at p. 309.

(a) Market Structure

The appellants and the AQPP have devoted a substantial part of their argument to a demonstration that no clear market-share guideline can be found in the cases. They have brought to the attention of this Court two works, W. T. Stanbury, Legislation to Control Agreements in Restraint of Trade in Canada: Review of the Historical Record and Proposals for Reform (1989), cited with approval in Association québécoise des pharmaciens propriétaires v. Canada (Procureur général), [1991] R.J.Q. 205 (Sup. Ct.), and Stanbury and G. B. Reschenthaler, "Reforming Canadian Competition Policy: Once More unto the Breach" (1981), 5 Can. Bus. L.J. 381, where the authors express doubts about the possibility of discerning a market-share threshold for liability in conspiracy cases. Indeed market share as such cannot suffice to conclude on the structure of the market, and s. 32(1)(c) would lose some of its effectiveness and would stray from its objectives if it incorporated a market-share threshold. Market share alone is not determinative, as was rightly pointed out in Canadian General Electric, at p. 501.

The aim of the market structure inquiry is to ascertain the degree of market power of the parties to the agreement, as was said in Canadian Coat & Apron Supply Ltd., at p. 64. In this respect, many factors other than market share are relevant. Some were listed in J. W. Mills, at pp. 307-8: (1) the number of competitors and the concentration of competition, (2) barriers to entry, (3) geographical distribution of buyers and sellers, (4) differences in the degree of integration among competitors, (5) product differentiation, (6) countervailing power and (7) cross-elasticity of demand (see also Canadian General Electric, supra). This list is not limitative: for instance, I note that in its 1984 Merger Guidelines, 49 Fed. Reg. 26823, the United States Department of Justice, Antitrust Division, proposed the ability to raise prices on a given product by five percent over a year without losses as the yardstick for market power. This approach may or may not be appropriate in the context of s. 32(1)(c) of the Act.

Market power is the ability to behave relatively independently of the market. This is precisely what s. 32(1)(c) of the Act seeks to prevent. As this Court has always held in its previous judgments, the aim of the Act is to secure for the Canadian public the benefit of free competition. Excessive market power runs against the objectives of the Act. When it occurs in the context of a conspiracy to restrict competition, s. 32(1)(c) will apply. It can be presumed that Parliament did not wish s. 32(1)(c) to apply in the absence of market power. Absent such power, agreements to restrict competition would either benefit the public by allowing small firms to consolidate their position and be more competitive, or dissolve under competitive pressures (see Dunlop et al., supra, at p. 114).

The level of market power necessary to trigger the application of s. 32(1)(c) is not necessarily the same as for other sections of the Act. For instance, s. 51 of the Act (now s. 79), prohibiting abuses of dominant position, contemplates at subs. (1)(a) that the holders of the dominant position "substantially or completely control, throughout Canada or any area thereof, a class or species of business". The required degree of market power under s. 51 of the Act comprises "control", and not simply the ability to behave independently of the market.

The application of s. 32(1)(c) of the Act does not presuppose such a degree of market power, as s. 32(1.1) clearly enunciates. Parties to the agreement need not have the capacity to influence the market. What is more relevant is the capacity to behave independently of the market, in a passive way. A moderate amount of market power is required to achieve this (see R. v. Abitibi Power & Paper Co. (1960), 1960 CanLII 501 (QC CQ), 131 C.C.C. 201 (Que. Q.B. (Crown side)), at pp. 249-52).

I note that the competition law of both the United States and the European Communities comprises an analogous requirement of minimal market power in cases of agreements to restrain competition.

In the United States, the nature of the impugned restraints will determine whether they are assessed according to a per se rule or a rule of reason. This operation has diverted much attention away from the substance of § 1 of the Sherman Act and, to correct this, the Supreme Court has to a certain extent blurred the distinction between the two approaches in NCAA v. Board of Regents of the University of Oklahoma, 468 U.S. 85 (1984). There the United States Supreme Court held that naked restrictions on price and output did not require a finding of market power, for their unreasonableness could be presumed (at pp. 100-101, 109-10). The possibility of per se rules allows for a presumption of unreasonableness (Canadian law does not offer this possibility, since the word "undue" appears at s. 32(1)(c) of the Act), but otherwise some showing of market power is necessary to evidence a genuine adverse effect on competition (See Federal Trade Commission v. Indiana Federation of Dentists, 476 U.S. 447 (1986), at pp. 460-61, and Areeda, supra, at ¶1503, pp. 376-77).

In the European Communities, the Commission has given the Notice of 3 September 1986 on agreements of minor importance which do not fall under Article 85(1) of the Treaty establishing the European Economic Community, O.J.E.C., 12 September 1986, No. C 231/2, and in addition the Court of Justice of the European Communities has long recognized that agreements which do not significantly affect the common market fall outside of the scope of Article 85 of the Treaty of Rome (Völk v. Établissements J. Vervaecke S.p.r.l., Case 5/69, [1969] E.C.R. 295, and S.A. Cadillon v. Firma Höss Maschinenbau K.G., Case 1/71, [1971] E.C.R. 351).

(b) Behaviour

The second part of the framework found in case law involves an examination of the behaviour of firms. In both R. v. McGavin Bakeries Ltd. (No. 6) (1951), 1951 CanLII 250 (AB SCTD), 3 W.W.R. (N.S.) 289 (Alta. S.C.), at p. 303, and Canadian General Electric, supra, at pp. 531-32, the court considered which facet of competition was affected by the agreement (price, quality, service or other) and whether it was the prime concern of the buying public. The object of the agreement is without doubt the most important behaviourial element in the inquiry, but others may be relevant, such as the manner in which the agreement has been or will be carried out and, in general, any behaviour that tends to reduce competition or limit entry (see J. W. Mills, supra, at p. 309).

The aim of the inquiry is the likely effect of the agreement. I can only quote from R. v. Northern Electric Co., 1955 CanLII 392 (ON SC), [1955] 3 D.L.R. 449 (Ont. H.C.), at p. 469:

In considering whether the agreement or conspiracy comes within the statute, one does not judge the unlawfulness by what was done pursuant to the agreement (although this may be evidence of the agreement) but . . . one examines the nature and scope of the agreement as proved and decides whether that agreement, if carried into effect, would prejudice the public interest in free competition to a degree that in fact would be undue.

See also Aetna Insurance, supra, at p. 747. The agreement must be or must have been likely to injure competition, irrespective of any actual effect it may have had (although evidence of the effects offers good guidance as to the likely effects of the agreement). Some agreements do not constitute behaviour likely to injure competition, while others are highly injurious.

The Act itself gives some guidance as to which behaviour may or may not be injurious to competition. Section 32(2) contains a list of permissible fields, lying outside the scope of s. 32(1)(c). Section 32(3) lists impermissible fields, to which s. 32(1)(c) will apply even if the agreement relates to one of the fields enumerated in s. 32(2). These lists bear some conceptual resemblance to the elaborate system of "black", "white" and "grey" clauses found in the regulations of the EC Commission exempting broad categories of agreements from Article 85 of the Treaty of Rome (see for instance Commission Regulation (EEC) No. 1983/83 of 22 June 1983 on the application of Article 85(3) of the Treaty to categories of exclusive distribution agreements, O.J.E.C., 30 June 1983, No. L 173/1, at arts. 1, 2). Given the relative smallness of the Canadian market when compared with the European Community, though, one cannot expect Canadian legislation to provide as much detail as EC regulations. Section 32(6) adds an exception for agreements on standards of competence and integrity in the field of services.

Section 32(1)(c) therefore requires, in addition to some market power, some behaviour likely to injure competition. It is the combination of the two that makes a lessening of competition undue. Many combinations are possible. For one, market power may come from the agreement. The agreement could either have an "internal" effect, in consolidating the market power of the parties (as is the case with price-fixing) or have an "external" effect, in weakening competition and thus increasing the market power of the parties (as is the case with market-sharing). Market power may also exist independently of the agreement, in which case any anti-competitive effect of the agreement will be suspicious. A particularly injurious behaviour may also trigger liability even if market power is not so considerable. These are only examples of possible combinations of market power and behaviour likely to injure competition that will be "undue" under s. 32(1)(c) of the Act.

In summary, I find that s. 32(1)(c) of the Act and its companion interpretative provision s. 32(1.1) do not violate s. 7 of the Charter on grounds of vagueness. The word "unduly" as such carries a connotation of seriousness. Considering further that s. 32(1)(c) of the Act is one of the oldest and most important parts of Canadian public policy in the economic field, and that it mandates a partial rule of reason inquiry into the seriousness of the competitive effects of the agreement, Parliament has sufficiently delineated the area of risk and the terms of debate to meet the constitutional standard. Moreover, the rest of the Act and the case law have outlined a process of examination of market structure and behaviour under s. 32(1)(c) of the Act, thus making it even more precise. I note that the ECHR has also found a comparable competition statute to be "prescribed by law" in Barthold, supra.

This holding is also dispositive of the further argument of the appellants on the alleged unconstitutionality of the indictment with respect to s. 11(a) of the Charter.

VI. The Mental Element of Section 32(1)(c) of the Act and Section 7 of the Charter

It may be helpful to set out once again the relevant provisions of the Act. Section 32(1) provides:

32. (1) Every one who conspires, combines, agrees or arranges with another person

(c) to prevent, or lessen, unduly, competition in the production, manufacture, purchase, barter, sale, storage, rental, transportation or supply of a product, or in the price of insurance upon persons or property,

is guilty of an indictable offence and is liable to imprisonment for five years or a fine of one million dollars or to both.

The Act was amended in 1976 by the addition of s. 32(1.1) which reads:

(1.1) For greater certainty, in establishing that a conspiracy, combination, agreement or arrangement is in violation of subsection (1), it shall not be necessary to prove that the conspiracy, combination, agreement or arrangement, if carried into effect, would or would be likely to eliminate, completely or virtually, competition in the market to which it relates or that it was the object of any or all of the parties thereto to eliminate, completely or virtually, competition in that market.

In 1986, further amendments to the Act (renamed the Competition Act) were adopted, including s. 32(1.3):

(1.3) For greater certainty, in establishing that a conspiracy, combination, agreement or arrangement is in contravention of subsection (1), it is necessary to prove that the parties thereto intended to and did enter into the conspiracy, combination, agreement or arrangement, but it is not necessary to prove that the parties intended that the conspiracy, combination, agreement or arrangement have an effect set out in subsection (1).

How then should those sections be interpreted in light of s. 7 of the Charter? The reasons of this Court in R. v. Vaillancourt, supra, at p. 652, make it clear that the requirement of a mental element, sometimes referred to as an element of fault, has now been raised from an interpretative presumption to a constitutional guaranty. This reasoning was based upon the principle that the morally innocent should not be punished. An element of fault must exist before punishment can be justified.

From the reasons of this Court in R. v. Wholesale Travel Group Inc., 1991 CanLII 39 (SCC), [1991] 3 S.C.R. 154, it can be seen that a minimum fault requirement with respect to every criminal or regulatory offence satisfies the requirements of s. 7. That same case indicates, at p. 238:

That fault may be demonstrated by proof of intent, whether subjective or objective, or by proof of negligent conduct, depending on the nature of the offence. . . .

. . . Mens rea focuses on the mental state of the accused and requires proof of a positive state of mind such as intent, recklessness or wilful blindness. Negligence, on the other hand, measures the conduct of the accused on the basis of an objective standard, irrespective of the accused's subjective mental state.

The sections of the Act set out above require the proof of two fault elements: one subjective, the other objective.

To satisfy the subjective element, the Crown must prove that the accused had the intention to enter into the agreement and had knowledge of the terms of that agreement. Once that is established, it would ordinarily be reasonable to draw the inference that the accused intended to carry out the terms in the agreement, unless there was evidence that the accused did not intend to carry out the terms of the agreement.

In order to satisfy the objective element of the offence, the Crown must establish that on an objective view of the evidence adduced the accused intended to lessen competition unduly. This surely does not impose too high a burden on the Crown. Section 32(1)(c) requires that the Crown demonstrate that the effect of the agreement will be to prevent competition or to lessen it unduly. Once again, it would be a logical inference to draw that a reasonable business person who can be presumed to be familiar with the business in which he or she engages would or should have known that the likely effect of such an agreement would be to unduly lessen competition. Thus in proving the actus reus that the agreement was likely to lessen competition unduly, the Crown could, in most cases, establish the objective fault element that the accused as a reasonable business person would or should have known that this was the likely effect of the agreement.

It must be remembered that if there are two possible interpretations of a statutory provision, one of which embodies the Charter values and the other does not, that which embodies the Charter values should be adopted. (See Hills v. Attorney General of Canada, 1988 CanLII 67 (SCC), [1988] 1 S.C.R. 513.)

In summary then, the Crown must establish the subjective fault elements that the accused had the intention to enter into the agreement and was aware of its terms. As well, the Crown must demonstrate that the proof, viewed objectively (i.e., by a reasonable business person), establishes that the accused was aware or ought to have been aware that the effect of the agreement entered into by the accused would be to prevent or lessen competition unduly.

Section 32(1)(c) of the Act does not therefore violate s. 7 of the Charter.

VII. Answers

I would answer the constitutional questions as follows:

1. Is s. 32(1)(c) of the Combines Investigation Act, R.S.C. 1970, c. C‑23 (now s. 45(1)(c) of the Competition Act, R.S.C., 1985, c. C‑34) in whole or in part inconsistent with s. 7 of the Canadian Charter of Rights and Freedoms?

Answer: No.

2. Is s. 32(1.1) of the Combines Investigation Act, R.S.C. 1970, c. C‑23 (now s. 45(2) of the Competition Act, R.S.C., 1985, c. C‑34) inconsistent with s. 7 of the Canadian Charter of Rights and Freedoms?

Answer: No.

3. If the answer to questions 1 or 2 is yes, is the infringement nevertheless justified under s. 1 of the Canadian Charter of Rights and Freedoms?

Answer: It is not necessary to answer this question.

VIII. Conclusion

I would dismiss the appeal.

Appeal dismissed.

Solicitors for the appellants: McInnes, Cooper & Robertson, Halifax; Patterson Kitz, Halifax.

Solicitor for the respondent: The Department of Justice (Legal Branch, Consumer & Corporate Affairs), Hull.

Solicitor for the intervener the Attorney General for Ontario: The Deputy Attorney General of Ontario, Toronto.

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

Solicitors for the interveners the Association québécoise des pharmaciens propriétaires et al.: McCarthy Tétrault, Montréal.