Judgments

[2000] 2 F.C. 117

A-68-96

(T-2257-93)

Richard Sauvé (Respondent) (Plaintiff)

v.

The Chief Electoral Officer of Canada, the Solicitor General of Canada, the Attorney General of Canada (Appellants) (Defendants)

(T-1084-94)

Sheldon McCorrister, Chairman, Lloyd Knezacek, Vice Chairman on their own behalf and on behalf of the Stony Mountain Institution Inmate Welfare Committee and Clair Woodhouse, Chairman, Aaron Spence, Vice Chairman on their own behalf and on behalf of the Native Brotherhood Organization of Stony Mountain Institution and Serge Bélanger, Emile A. Bear and Randy Opoonechaw (Respondents) (Plaintiffs)

v.

The Attorney General of Canada (Appellant) (Defendant)

Indexed as: Sauvé v. Canada (Chief Electoral Officer) (C.A.)

Court of Appeal, Isaac C.J., Desjardins and Linden JJ.A.—Ottawa, June 8, 9 and October 21, 1999.

Constitutional law — Charter of Rights — Democratic rights — Respondents challenging constitutionality of Canada Elections Act, s. 51(e) — Crown conceding impugned provision violates Charter, s. 3 — Whether provision saved by Charter, s. 1 — Oakes test applied — Objectives of legislation sufficiently pressing, substantial to warrant infringement of Charter right — Legislation rationally connected to intended objective — S. 51(e) impairing Charter right in appropriately minimal way — Parliament not bound to use least restrictive means to achieve legislative objectives — Legislation targeted to disqualify most serious offenders from voting — Reasonable limit demonstrably justified in free, democratic society under Charter, s. 1.

Constitutional law — Charter of Rights — Equality rights — Respondents seeking declaration Canada Elections Act, s. 51(e) contrary to Charter, s. 15 — Status of prisoner not analogous ground for purposes of Charter, s. 15 — Not personal characteristic immutable, changeable only at unacceptable cost to personal identity — Prisoners not constituting analogous group warranting protection under Charter’s equality provision — Canada Elections Act, s. 51(e) not violating Charter, s. 15(1).

Elections — Canada Elections Act, s. 51(e), as amended, disqualifying from voting only prisoners serving sentences of two years or more — Whether provision meets minimal impairment, proportionality tests mandated by Charter, s. 1 — History of prisoner disenfranchisement in pre-Charter, post-Charter eras reviewed — Former s. 51(e) invalidated by S.C.C. as drawn too broadly — Parliament seeking to enact new law in conformity with Charter, case law — Provision being of hybrid criminal, electoral nature — Tailored to affect only most serious offenders — Proportionality between deleterious, salutary effects of measure.

Penitentiaries — Respondents disqualified from voting at federal election as serving sentence of two years or more — Canada Elections Act, s. 51(e) denying vote only to those serving two years or more, not to all of those in penal institutions — Parliament entitled to add civil consequences to criminal sanction — Deleterious effects of impugned provision proportional to significance of objectives, beneficial effects of measure — Context of each case paramount — Canada Elections Act, s. 51(e) infringing Charter, s. 3, but saved by s. 1.

These were an appeal and a cross-appeal from a Trial Division decision striking down paragraph 51(e) of the Canada Elections Act on the ground that it violates section 3 of the Canadian Charter of Rights and Freedoms and could not be saved by section 1 of the Charter. The respondent, Sauvé, who had been convicted of first degree murder and sentenced to 25 years’ incarceration, sought a declaration that paragraph 51(e) of the Act is contrary to sections 3 and 15 of the Charter. He is now on parole. The other respondents, who are Aboriginal inmates at the Stony Mountain Institution in Manitoba, filed a statement of claim seeking the same relief. In two previous cases, the Ontario Court of Appeal and the Federal Court of Appeal declared unconstitutional paragraph 51(e) of the Canada Elections Act, which provided, in its former version, that all inmates serving sentence in any penal institution for the commission of any offence were not qualified to vote at an election. These decisions were upheld by the Supreme Court of Canada which ruled that paragraph 51(e) was drawn too broadly and failed to meet the proportionality test. Shortly after, paragraph 51(e) was amended, so that only prisoners serving a sentence of two years or more in a correctional institution were prohibited from voting at a federal election. The main issue on appeal was whether paragraph 51(e) of the Act can be saved by section 1 of the Charter, more precisely whether it meets the minimal impairment and proportionality tests mandated by section 1 of the Charter.

Held (Desjardins J.A. dissenting), the appeal should be allowed, the cross-appeal should be dismissed.

Per Linden J.A.: In a section 1 analysis, context matters. Concepts such as minimal impairment and proportionality must be viewed in the light of the case before the Court. The first segment of the Oakes test is whether the objectives of the legislation are sufficiently pressing and substantial to warrant compromise of a Charter right. The two objectives of paragraph 51(e) of the Canada Elections Act, as submitted by the Crown and accepted by the Trial Judge, are: (a) the enhancement of civil responsibility and respect for the rule of law and (b) the enhancement of the general purposes of the criminal sanction. The government has proven on a balance of probabilities that the objectives it contends for indeed motivated the passage of this legislation and the Trial Judge made no error in accepting them. The federal government enacted this legislation both as an exercise of its criminal law power and an exercise of its right to legislate about electoral law. The objectives of enhancement of civic responsibility and respect for the rule of law and enhancement of the penal sanction are sufficiently pressing and substantial as to warrant an infringement of a Charter right. Fostering civic responsibility and respect for the rule of law is important enough, in some cases, to warrant compromise of Charter rights. The penal sanction may be utilized in ways that may compromise Charter rights.

The first requirement of the Oakes test being met, the second requirement involves a proportionality test. In order to show that legislation which violates a Charter right is saved by section 1, the government must demonstrate that there is a rational connection between the objectives which it seeks to further and the means by which it has chosen to do so. Legislation will be rationally connected to its intended objective if it (a) is designed to meet its objective, (b) is not arbitrary and (c) is based on assumptions which, logically applied, further the objective. The impugned legislation is rationally connected to the government’s first objective which is meant to foster civic responsibility and enhance respect for the rule of law. There is also a rational connection between the disenfranchisement of prisoners and the objective of furthering the criminal sanction. In the result, the legislation is rationally connected to the pressing and substantial objectives which motivated it. The second requirement of the Oakes test requires that the impugned state action impair a Charter right in an appropriately minimal way. Parliament need not, however, adopt the absolutely least-intrusive means of attaining its objectives. In assessing minimal impairment, a court must accord some deference to Parliament. The context of each case must determine the level of deference given by the courts to Parliament’s impairment of the Charter right. It is Parliament’s role to maintain and enhance the integrity of the electoral process. Parliament’s choices regarding the gravity of offences and the punishment which follows from conviction are entitled to considerable deference. The Trial Judge erred in reasoning that judge-imposed disenfranchisement would be a less intrusive manner of achieving the legislative objective than the current legislation. This law was carefully tailored to affect only Canada’s most serious offenders. The disqualification applies only to those actually incarcerated after being sentenced to penitentiary terms of two years or more. This legislation is not exclusively an attempt to add a Parliamentary sentence to that imposed by the judiciary. In addition to its electoral component, it creates a consequence to criminal conviction, but that does not transform it into a sentencing provision. It is well settled that legislation may validly provide for a civil disability arising out of a criminal conviction. Civil consequences of a criminal act are not necessarily viewed as part of the criminal sentence. The prohibition contained in this legislation is hybrid in that it possesses elements of the criminal sanction as well as elements of civil disability based on electoral law. Parliament, basing itself on electoral policy, is entitled to add civil consequences to the criminal sanction in subtle, multi-dimensional ways.

The third part of the proportionality test is whether the deleterious effects of the measure are proportional to the significance of the objectives and the beneficial effects of the measure. The context of each case is always paramount and there can be no rigid test for an assessment of proportionality. In addition to electoral considerations, the main motivations in passing this law were the retributive and denunciatory aspects of the penal sanction. The courts cannot prevent Parliament from proportionately compromising Charter rights in the name of denouncing crime, even if they disagree with Parliament’s penal philosophy. There are salutary effects of this legislation as well as valid objectives which were identified to the Court. Its main salutary effect is to express the sense of societal values of the community in relation to serious criminal behaviour and the right to vote in our society. It sends a message signalling Canadian values to the effect that those people who are found guilty of the most serious crimes will, while separated from society, lose access to one of the levers of electoral power. This legislation proclaims that values of civic responsibility are important to Canadians. The signal itself is an important benefit of the law. Moreover, disenfranchisement is a meaningful sanction which is noticed by offenders. Lastly, this legislation can be seen as a gentler, more humane alternative to additional incarceration. In the battle against crime, Courts cannot limit Parliament to a single punitive tool. On the other hand, the sole deleterious effect of the legislation is the withdrawal of the Charter-guaranteed right to vote. While this deprivation is serious, several facts were brought to the attention of the Court which mitigate its deleterious nature. Viewed as a civil consequence imposed as an alternative to additional incarceration which attaches to the most serious sentences for the most serious crimes, it must be concluded that this measure is proportional.

The last issue was whether paragraph 51(e) of the Canada Elections Act infringes section 15 of the Charter. The recent Supreme Court of Canada decision in Law v. Canada (Minister of Employment and Immigration) is now the starting point for any analysis of discrimination under subsection 15(1) of the Charter. It must first be shown that the impugned legislation or state action draws a distinction between the claimant and others, second that the distinction has been drawn on an enumerated or analogous ground and third, that the distinction drawn on the basis of an enumerated or analogous ground is discriminatory in its nature. The status of prisoner does not constitute an analogous ground for purposes of section 15 of the Charter. It cannot be described as a personal characteristic that is immutable or changeable only at unacceptable cost to personal identity. Prisoners per se do not constitute a group analogous to those enumerated in subsection 15(1) of the Charter. While paragraph 51(e) of the Act violates section 3 of the Charter, it is a reasonable limit demonstrably justified in a free and democratic society within the meaning of section 1 of the Charter.

Per Desjardins J.A. (dissenting): Paragraph 51(e) of the Canada Elections Act still carries with it some reminiscence of outdated stereotypes concerning the social status of prisoners. Depriving prisoners of the right to vote is an anachronistic concept. The dominant character of this legislation is the further punishment of the detainee following the commission of serious crimes. Although the link between the crime and the electoral system is somewhat remote, the penal character of the legislation is the focal point. An infringement of a fundamental right or freedom is only justified under section 1 of the Charter when both components of the Oakes test are satisfied. The test is to be applied flexibly and in context. The first branch of the Oakes test is whether the objectives of the impugned legislation are sufficiently pressing and substantial to justify overriding a protected right. The new version of paragraph 51(e) involves the state interfering in the fundamental rights of a historically vulnerable group of people for a punitive purpose on behalf of society at large. It is a case where the state behaves as a “singular antagonist”. The justifications advanced by the appellants with regard to the voting process were highly abstract. Their experts’ testimony established only that there might be some possible benefits from the legislation over the long term. They did not establish that the legislation was curing any problems or that its absence would result in any tangible harm. Benefits and harms must be reasonably established, in the present, before Parliament can rely upon them as its rationale for infringing constitutionally protected rights. In light of the abstract and uncertain benefits to be obtained from paragraph 51(e) of the Act, the Trial Judge’s finding that the objectives of the impugned legislation were in answer to a pressing and substantial concern should be accepted with some reservation.

The second branch of the Oakes test comprises three parts. First, there must be a rational connection between the means chosen and the goals chosen by the legislation. The Trial Judge noted the absence of empirical evidence which would suggest that the disenfranchisement of prisoners actually serves a morally educative function. The absence of a rational connection between disenfranchisement and serving the objectives of rehabilitation could not be overlooked. The enhancement of the criminal law and the sending of a morally educative message are the only objectives rationally connected to the goals pursued by the legislation. Second, the measures must impair constitutional rights and freedoms “as little as possible”. The legislation in question, not being tailored to the goals pursued by it, fails the minimal impairment test. It remains a general form of ostracism or banishment even with a marker of two years or more in prison. The sentencing judge would be better suited to pronounce disenfranchisement as a further punishment. Automatic disenfranchisement is excessive in character. The sentencing judge might apply the sanction sparingly, but the process would be less arbitrary and more significant. The Trial Judge correctly applied the minimum impairment test. This is not a situation where the courts are substituting their views for that of Parliament. It is a case where the Charter plays a role in framing the legislation. Third, there must be proportionality between the deleterious effects of the measures which are responsible for infringing the Charter right and the objectives of the legislation, and there must be proportionality between the deleterious and salutary effects of those measures. Most of the evidence offered by the appellants to support the impugned legislation amounted to an educated guess rather than the clear demonstration required by section 1 of the Charter. The statements made by the appellants’ witnesses were not the kind of evidence required to support a section 1 analysis. The impugned legislation, being a last remnant of exclusionary practices, cannot resist the Charter and is therefore invalid.

STATUTES AND REGULATIONS JUDICIALLY CONSIDERED

Canada Elections Act, R.S.C. 1970 (1st Supp.), c. 14, s. 14(4)(e).

Canada Elections Act, R.S.C., 1985, c. E-2, ss. 14(4), 51(e) (as am. by S.C. 1993, c. 19, s. 23), 77(1) (as am. idem, s. 34).

Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44], ss. 1, 3, 12, 15.

Constitutional Act, 1791, 31 Geo. III, c. 31 (U.K.) [R.S.C., 1985, Appendix II, No. 3], s. XXIII.

Constitution Act, 1867, 30 & 31 Vict., c. 3 (U.K.) (as am. by Canada Act 1982, 1982, c. 11 (U.K.), Schedule to the Constitution Act, 1982, Item 1) [R.S.C., 1985, Appendix II, No. 5].Criminal Code, R.S.C., 1985, c. C-46, s. 748(1) (as am. by S.C. 1995, c. 22, s. 6).

Divorce Act, S.C. 1967-68, c. 24.

Election Act, R.S.B.C. 1996, c. 106, s. 30(b).

Elections Act, C.C.S.M., c. E-30.

Elections Act, R.S.N.B. 1973, c. E-3.

Electoral Franchise Act (The), S.C. 1885, c. 40, s. 3(1).

Federal Court Act, R.S.C., 1985, c. F-7, s. 57(5) (as am. by S.C. 1990, c. 8, s. 19).

Franchise Act, 1898 (The), S.C. 1898, c. 14, s. 6(4).

Indian Act, R.S.C., 1985, c. I-5, s. 78(1).

Motor Vehicle Act, R.S.B.C. 1979, c. 288.

Parliament of Canada Act, R.S.C., 1985, c. P-1, s. 41(2).

Public Service Superannuation Act, R.S.C., 1985, c. P-36, s. 42(1).

CASES JUDICIALLY CONSIDERED

FOLLOWED:

The Queen v. Oakes, [1986] 1 S.C.R. 103; (1986), 26 D.L.R. (4th) 200; 24 C.C.C. (3d) 321; 50 C.R. (3d) 1; 19 C.R.R. 308; 65 N.R. 87; 14 O.A.C. 335; Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497; (1999), 170 D.L.R. (4th) 1; 236 N.R. 1.

APPLIED:

Belczowski v. Canada, [1991] 3 F.C. 151 (1991), 5 C.R. (4th) 218; 6 C.R.R. (2d) 345; 42 F.T.R. 98 (T.D.); Egan v. Canada, [1995] 2 S.C.R. 513; (1995), 124 D.L.R. (4th) 609; 95 CLLC 210-025; 29 C.R.R. (2d) 79; 182 N.R. 161; 12 R.F.L. (4th) 201; Committee for the Commonwealth of Canada v. Canada, [1991] 1 S.C.R. 139; (1991), 77 D.L.R. (4th) 385; 4 C.R.R. (2d) 60; 120 N.R. 241; R. v. Butler, [1992] 1 S.C.R. 452; (1992), 89 D.L.R. (4th) 449; [1992] 2 W.W.R. 577; 70 C.C.C. (3d) 129; 11 C.R. (4th) 137; 8 C.R.R. (2d) 1; 78 Man. R. (2d) 1; 134 N.R. 81; 16 W.A.C. 1; Jackson v. Joyceville Penitentiary, [1990] 3 F.C. 55 (1990), 55 C.C.C. (3d) 50; 75 C.R. (3d) 174; 1 C.R.R. (2d) 327; 32 F.T.R. 96 (T.D.); Olson v. Canada, [1996] 2 F.C. 168 (1996), 107 F.T.R. 81 (T.D.); Alcorn v. Canada (Commissioner of Corrections), [1999] F.C.J. No. 330 (T.D.) (QL); Corbiere v. Canada (Minister of Indian and Northern Affairs), [1999] 2 S.C.R. 203; (1999), 173 D.L.R. (4th) 1; [1999] 3 C.N.L.R. 19; 239 N.R. 1.

CONSIDERED:

Sauvé v. Canada (Attorney General) (1992), 7 O.R. (3d) 481; 89 D.L.R. (4th) 644; 55 O.A.C. 219 (C.A.); Belczowski v. Canada, [1992] 2 F.C. 440 (1992), 90 D.L.R. (4th) 330; 12 C.R. (4th) 219; 9 C.R.R. (2d) 14; 132 N.R. 183 (C.A.); Sauvé v. Canada (Attorney General), [1993] 2 S.C.R. 438; (1993), 15 C.R.R. (2d) 1; 153 N.R. 242; 64 O.A.C. 124; Harvey v. New Brunswick (Attorney General), [1996] 2 S.C.R. 876; (1996), 178 N.B.R. (2d) 161; 137 D.L.R. (4th) 142; 454 A.P.R. 161; 37 C.R.R. (2d) 189; 201 N.R. 1; RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; (1995), 127 D.L.R. (4th) 1; 100 C.C.C. (3d) 449; 62 C.P.R. (3d) 417; 31 C.R.R. (2d) 189; 187 N.R. 1; M. v. H., [1999] 2 S.C.R. 3; (1999), 171 D.L.R. (4th) 577; 238 N.R. 179; 21 O.A.C. 1 (S.C.C.); Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143; (1989), 56 D.L.R. (4th) 1; [1989] 2 W.W.R. 289; 34 B.C.L.R. (2d) 273; 25 C.C.E.L. 255; 10 C.H.R.R. D/5719; 36 C.R.R. 193; 91 N.R. 255; Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927; (1989), 58 D.L.R. (4th) 577; 25 C.P.R. (3d) 417; 94 N.R. 167; Lévesque v. Canada (Attorney General), [1986] 2 F.C. 287 (1985), 25 D.L.R. (4th) 184; 20 C.R.R. 15 (T.D.); Badger v. Canada (Attorney General) et al. (1988), 55 Man. R. (2d) 211 (Q.B.); Badger v. Canada (Attorney General) (1988), 55 D.L.R. (4th) 177; [1989] 1 W.W.R. 216; 55 Man. R. (2d) 198 (C.A.); leave to appeal to S.C.C. refused, [1989] 1 S.C.R. v; Jolivet and Barker and The Queen and Solicitor General of Canada, Re (1983), 1 D.L.R. (4th) 604 (B.C.S.C.); Attorney General of Canada v. Gould, [1984] 1 F.C. 1133; (1984), 13 D.L.R. (4th) 485; 42 C.R. (3d) 88; 54 N.R. 232 (C.A.); Sauvé v. Canada (Attorney General) (1988), 66 O.R. (2d) 234; 53 D.L.R. (4th) 595 (H.C.); Thomson Newspapers Co. v. Canada (Attorney General), [1998] 1 S.C.R. 877; (1998), 38 O.R. (3d) 735; 159 D.L.R. (4th) 385; 226 N.R. 1; 109 O.A.C. 201; Lavoie v. Canada, [2000] 1 F.C. 3 (1999), 174 D.L.R. (4th) 588 (C.A.); R. v. Goltz, [1991] 3 S.C.R. 485; (1991), 11 W.A.C. 161; 61 B.C.L.R. (2d) 145; 5 B.C.A.C. 161; 67 C.C.C. (3d) 481; 8 C.R. (4th) 82; 7 C.R.R. (2d) 1; 31 M.V.R. (2d) 137; 131 N.R. 1; R. v. M. (C.A.), [1996] 1 S.C.R. 500; (1996), 73 B.C.A.C. 81; 105 C.C.C. (3d) 327; 46 C.R. (4th) 260; 194 N.R. 321; Provincial Secretary of Prince Edward Island v. Egan, [1941] S.C.R. 396; (1941), 3 D.L.R. 305; 76 C.C.C. 227; Armstrong v. R., [1996] 1 C.T.C. 2745 (T.C.C.).

REFERRED TO:

Byatt et al. v. Alberta et al. (1998), 216 A.R. 100; 158 D.L.R. (4th) 644; 62 Alta. L.R. (3d) 10 (C.A.); Ross v. New Brunswick School District No. 15, [1996] 1 S.C.R. 825; (1996), 133 D.L.R. (4th) 1; 37 Admin. L.R. (2d) 131; 195 N.R. 81; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624; (1997), 151 D.L.R. (4th) 577; 96 B.C.A.C. 81; 218 N.R. 161; Miron v. Trudel, [1995] 2 S.C.R. 418; (1995), 124 D.L.R. (4th) 693; 29 C.R.R. (2d) 189; [1995] I.L.R. 1-3185; 10 M.V.R. (3d) 151; 181 N.R. 253; 81 O.A.C. 253; 13 R.F.L. (4th) 1; Driskell v. Manitoba (Attorney General), [1999] M.J. No. 352 (Q.B.) (QL); United Food and Commercial Workers, Local 1518 (U.F.C.W.) v. KMart Canada Ltd., [1999] S.C.J. No. 44 (QL); Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835; (1994), 120 D.L.R. (4th) 12; 94 C.C.C. (3d) 289; 34 C.R. (4th) 269; 25 C.R.R. (2d) 1; 175 N.R. 1; 76 O.A.C. 81; McKinney v. University of Guelph, [1990] 3 S.C.R. 229; (1990), 76 D.L.R. (4th) 545; 91 CLLC 17,004; 2 C.R.R. (2d) 1; 118 N.R. 1; 45 O.A.C. 1; Stoffman v. Vancouver General Hospital, [1990] 3 S.C.R. 483; [1991] 1 W.W.R. 577; (1990), 52 B.C.L.R. (2d) 1; 91 CLLC 17,003; Vriend v. Alberta, [1998] 1 S.C.R. 493; (1998), 212 A.R. 237; 156 D.L.R. (4th) 385; [1999] 5 W.W.R. 451; 67 Alta. L.R. (3d) 1; 224 N.R. 1; Reynolds v. A.G.B.C. (1984), 53 B.C.L.R. 394 (C.A.); Badger et al. v. Attorney General of Manitoba (1986), 30 D.L.R. (4th) 108; 39 Man. R. (2d) 107; 27 C.C.C. (3d) 158; 51 C.R. (3d) 163; 21 C.R.R. 277 (Q.B.); Grondin v. Ontario (Attorney General) (1988), 65 O.R. (2d) 427 (H.C.); Gould v. Attorney General of Canada, [1984] 1 F.C. 1119; (1984), 42 C.R. (3d) 78 (T.D.); Gould v. Attorney General of Canada et al., [1984] 2 S.C.R. 124; (1984), 13 D.L.R. (4th) 485; 42 C.R. (3d) 88; 53 N.R. 394; R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 713; (1986), 35 D.L.R. (4th) 1; 30 C.C.C. (3d) 385; 87 CLLC 14,001; 55 C.R. (3d) 193; 28 C.R.R. 1; 71 N.R. 161; 19 O.A.C. 239; United States of America v. Cotroni; United States of America v. El Zein, [1989] 1 S.C.R. 1469; (1989), 23 Q.A.C. 182; 96 N.R. 321; 48 C.C.C. (3d) 193; R. v. Laba, [1994] 3 S.C.R. 965; (1994), 120 D.L.R. (4th) 175; 94 C.C.C. (3d) 385; 34 C.R. (4th) 360; 25 C.R.R. (2d) 92; 174 N.R. 321; 76 O.A.C. 241; R. v. Guiller (1986), 48 C.R. (3d) 226 (Ont. Dist. Ct.); R. v. Luxton, [1990] 2 S.C.R. 711; (1990), 111 A.R. 161; [1990] 6 W.W.R. 137; 76 Alta. L.R. (2d) 43; 58 C.C.C. (3d) 449; 79 C.R. (3d) 193; 50 C.R.R. 175; 112 N.R. 193; Condo v. Ontario (Registrar of Motor Vehicles), [1999] O.J. No. 1601 (Div. Ct.) (QL); R. v. Joslin (1981), 59 C.C.C. (2d) 512; 10 M.V.R. 29 (Ont. C.A.); Ross v. Registrar of Motor Vehicles et al., [1975] S.C.R. 5; (1973), 42 D.L.R. (3d) 68; 14 C.C.C. (2d) 322; 23 C.R.N.S. 319; 1 N.R. 9; Barclay (D. & G.) Builders Ltd. and St. Jane Plaza Ltd., Re, [1973] 1 O.R. 579; (1973), 31 D.L.R. (3d) 675 (Div. Ct.); R. v. Mitri, [1989] O.J. No. 1873 (Prov. Ct.) (QL); Canadian Broadcasting Corp. v. New Brunswick (Attorney General), [1996] 3 S.C.R. 480; (1996), 182 N.B.R. (2d) 81; 139 D.L.R. (4th) 385; 463 A.P.R. 81; 110 C.C.C. (3d) 193; 2 C.R. (5th) 1; 203 N.R. 169; Pelech v. Pelech, [1987] 1 S.C.R. 801; (1987), 38 D.L.R. (4th) 641; [1987] 4 W.W.R. 481; 14 B.C.L.R. (2d) 145; 17 C.P.C. (2d) 1; 76 N.R. 81; 7 R.F.L. (3d) 225; R. v. Stewart (1991), 27 M.V.R. (2d) 187 (B.C.S.C.); Eaton v. Brant County Board of Education, [1997] 1 S.C.R. 241; (1997), 31 O.R. (3d) 574; 142 D.L.R. (4th) 385; 207 N.R. 171; Batchewana Indian Band (Non-resident members) v. Batchewana Indian Band, [1997] 1 F.C. 689 (1996), 142 D.L.R. (4th) 122; [1997] 3 C.N.L.R. 21; 206 N.R. 85 (C.A.); McKinnon (R.J.) v. M.N.R., [1991] 2 C.T.C. 2284; (1991), 91 DTC 1002 (T.C.C.); Mulligan v. Canada, [1996] T.C.J. No. 1688 (T.C.C.) (QL); Wells v. R., [1997] 1 C.T.C. 2112 (T.C.C.).

AUTHORS CITED

Beaudoin, Gérald A. and Errol P. Mendes, eds. The Canadian Charter of Rights and Freedoms, 3rd ed. Toronto: Carswell, 1996.

Canada. House of Commons. Special Committee on Electoral Reform. Minutes of Proceedings and Evidence, Issue No. 10, 1993.

Hampton, Jean. “Punishment, Feminism, and Political Identity: A Case Study in the Expressive Meaning of the Law” (1998), 11 Can. J. Law & Jur. 23.

Hogg, P. W. and A. A. Bushell. “The Charter Dialogue Between Courts and Legislatures” (1997), 35 Osgoode Hall L. J. 75.

House of Commons Debates, Vol. 14, 3rd Sess., 34th Parl., 1993.

Landreville, P. and L. Lemonde. “Voting Rights for Prison Inmates” in Royal Commission on Electoral Reform and Party Financing. Democratic Rights and Electoral Reform in Canada (Research Studies; 10). Toronto: Dundurn Press, 1991.

MLA Committee Making Recommendations on Restrictions on Prisoner Voting in the Alberta Election Act. Promoting Responsible Citizenship: Report to the Minister of Justice. Edmonton: The Committee, 1998.

Mewett & Manning on Criminal Law, 3rd ed. by A. W. Mewett and M. Manning. Markham, Ont.: Butterworths, 1994.

Morley, Gareth. “A Just Measure of Pain? Sentencing and Sentencing Reform in the Era of the Charter” (1997), 55 U.T. Fac. L. Rev. 269.

Ruby, Clayton. Sentencing, 2nd ed. Toronto: Butterworths, 1980.

APPEAL and CROSS-APPEAL from a Trial Division decision ([1996] 1 F.C. 857 (1995), 132 D.L.R. (4th) 136; 106 F.T.R. 241) striking down paragraph 51(e) of the Canada Elections Act on the ground that it violates section 3 of the Charter and could not be saved by section 1. Appeal allowed, cross-appeal dismissed.

APPEARANCES:

David G. Frayer, Q.C., and Gérald L. Chartier for appellants.

Arne Peltz for respondents McCorrister et al.

Fergus J. O’Connor for respondent Richard Sauvé.

Gerald D. Chipeur for intervener Attorney General of Alberta.

SOLICITORS OF RECORD:

Deputy Attorney General of Canada for appellants.

Public Interest Law Centre, Winnipeg, for respondents McCorrister et al.

Fergus J. O’Connor, Kingston, Ontario, for respondent Richard Sauvé.

Fraser Milner, Calgary, for intervener Attorney General of Alberta.

The following are the reasons for judgment rendered in English by

[1]        Desjardins J.A. (dissenting): This case brings before us, for the second time, the issue of the rights of prisoners to vote in federal elections. An earlier federal statute, which disenfranchised all prisoners for the purposes of federal election, was declared invalid by the Supreme Court of Canada. The law was then amended. The legislation at issue here, paragraph 51(e) of the Canada Elections Act[1] (the CEA or the Act), stipulates that prisoners serving sentences of two years or more in correctional institutions are prohibited from voting in federal elections. The provision, which came into force on May 6, 1993, reads:

51. The following persons are not qualified to vote at an election and shall not vote at an election:

…

(e) every person who is imprisoned in a correctional institution serving a sentence of two years or more.

[2]        The respondents contend that this legislation unduly infringes the right of prison inmates to vote in federal elections, a right which is guaranteed by section 3 of the Canadian Charter of Rights and Freedoms [being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44]] (the Charter). Section 3 provides that:

3. Every citizen of Canada has the right to vote in an election of members of the House of Commons or of a legislative assembly and to be qualified for membership therein.

[3]        The appellants, represented by the Attorney General of Canada, have admitted that this disenfranchisement constitutes a prima facie breach of section 3 of the Charter. The sole issue in this appeal is, therefore, essentially whether the disenfranchisement is justified under section 1 of the Charter which provides:

1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

[4]        The respondents have cross-appealed. They claim that paragraph 51(e) of the Act is in violation of section 15 of the Charter.

[5]        I shall deal first with the appeal and then with the cross-appeal.

A.        The appeal

[6]        The facts of this case have been set out by my colleague Linden J.A. who has also done a brief history of prisoner disenfranchisement legislation in pre-Charter Canada and a review of the case law on this matter following the adoption of the Charter. This allows me to undertake immediately an analysis of the law. But, before I do so, I wish to clarify that when the appeals of Sauvé v. Canada (Attorney General)[2] and Belczowski v. Canada[3] were heard together by the Supreme Court of Canada in Sauvé v. Canada (Attorney General)[4] [hereinafter] Sauvé No. 1 the Supreme Court dismissed both appeals because in the opinion of the Court:

… s. 51(e) is drawn too broadly and fails to meet the proportionality test, particularly the minimal impairment component of the test, as expressed in the s. 1 jurisprudence of the Court. [My emphasis.]

[7]        I observe that the Supreme Court of Canada, in that case, expressed concern with the proportionality test, particularly with the minimal impairment component of the test.

[8]        It is with this background in mind that I approach the case at bar.

I.          The nature and purpose of paragraph 51(e) of the Act

[9]        In order to determine whether paragraph 51(e) of the Act satisfies the requirements of section 1 of the Charter, it is first necessary to determine the nature and purpose of this provision.

[10]      Before the Trial Judge[5] and before us, the appellants submitted that the purposes of this legislation are to:

(1) enhance civic responsibility and respect for the rule of law; and

(2) enhance the general purposes of the criminal sanction.

[11]      The appellants submitted that the first objective of paragraph 51(e) of the Act is to cultivate among Canadian citizens an appreciation for, and an understanding of, the relationship between individuals’ rights and societal responsibilities; good citizenship, in particular. They argued that the impugned provision seeks to denounce citizens who have, through their serious criminal conduct, acted in a grossly disrespectful manner toward the lives, property, or dignity of their fellow citizens. This denunciatory message in the legislation aims at affirming the connection between participation in the voting process and commitment to the rule of law.

[12]      The appellants submitted that the second objective of the legislation is the enhancement of the general purposes of the criminal sanction. In their view, the suspension of an inmate’s right to vote until he has been released from prison sends a message which has a retributive, denunciatory and morally educative function.

[13]      The respondents argued that the disenfranchisement of prisoners serves no useful purpose at any level and that it is merely a remnant of the exclusionary practices of the past. They pointed to the marginalization of much of society in the past through property and gender qualifications for voting.[6] They noted that the idea that prisoners should be banished from society and suffer the loss of all of their civil rights has fallen by the wayside along with these other exclusionary practices. They submitted that the two-year cutoff was arbitrary and chosen for administrative convenience[7] and that the Crown’s expert witnesses had been unable to establish any actual benefit derived by society as a result of the disenfranchisement of prisoners.[8]

[14]      The Trial Judge carefully studied the history of the legislation in order to ascertain the grounds on which the impugned legislation was based.

[15]      He observed that, in 1993, the Act in force at the time of Sauvé No. 1 and Belczowski[9] was substantially amended by Bill C-114,[10] and that, prior to that, in November 1989, a Royal Commission on Electoral Reform and Party Financing (the Lortie Commission) was established for the purpose of inquiring into the appropriate principles and processes that should govern, inter alia, the election of members of the House of Commons. The Parliamentarians who voted on the amendment to the Act were thus well aware of the research studies and report of the Lortie Commission. The Trial Judge wrote:[11]

As is typical of royal commissions, numerous research studies accompanied the Lortie Commission’s recommendations. The prisoner disqualification, in particular, was addressed in an article by P. Landreville and L. Lemonde entitled “Voting Rights for Prisoner Inmates” [in Democratic Rights and Electoral Reform in Canada, Volume 10 of the Research Studies], in which the authors recommended that all prisoners be granted the right to vote. The Lortie Commission did not accept that recommendation; rather, it concluded that persons who had been convicted of an offence punishable by a maximum of life imprisonment, and who had been sentenced to a prison term of ten years or more, should be disqualified from voting during the period of incarceration (Recommendation 1.2.7.). Obviously, Parliamentarians had full access to the research studies, as well as the Final Report of the Royal Commission.

[16]      An intensive review by a Special Committee on Electoral Reform occurred in the House of Commons. The Trial Judge stated to that effect:[12]

The Minutes of the Proceedings of the Special Committee on Electoral Reform reveal Parliament’s concern that the courts should defer to Parliament on the issue of the prisoner disqualification. The Special Committee spent a great deal of time trying to determine whether a two-year limit for the disqualification was appropriate, or whether a cutoff of five years, or seven years, or ten years (as recommended by the Lortie Commission) was more justifiable. Eventually, the Special Committee recommended a two-year cutoff since, in their view, serious offenders may be considered to be those individuals who have been sentenced to a term of two years or more in a correctional institution. Generally, that means a federal penitentiary, but not exclusively. [My emphasis.]

[17]      The Trial Judge noted from the Minutes of the Proceedings of the Special Committee that the objective of enhancing punishment through the disenfranchisement of prisoners was clearly considered by the Committee.[13] For instance, during the discussions of the Committee, it was expressed that there must be some limits on the right to vote and that punishment encompass the disenfranchisement of all incarcerated offenders. The Trial Judge implied from comments drawn from the House of Commons Debates:[14]

… that the Government intended the prisoner disenfranchisement provision to have an educative effect.

[18]      Similarly, he wrote:[15]

… it was expressed that persons who choose to act against society must suffer the consequences, including the denial of freedom and the loss of privileges which free and responsible citizens enjoy“one of which is the right to vote. [My emphasis.]

[19]      He further wrote that the application of paragraph 51(e) was not offender-specific in the sense that it did not take into account the particular circumstances of the inmate before and after incarceration. It was sentence-specific. He was satisfied that sentences of two years or more involved serious crimes that reflected conduct which a court had determined to be sufficiently distasteful to have warranted such a sentence. He said:[16]

In the context of the considerable amount of study and discussion surrounding prisoner disenfranchisement before Bill C-114 was passed, I find a clear concern by Parliament regarding the characteristics of civic responsibility and respect for the rule of law in Canadian society. Furthermore, there is evidence of an intent to punish individuals who commit serious antisocial acts.

Specifically, the legislative text, in conjunction with the proceedings of the Special Committee on Electoral Reform, reveal that the provision is clearly directed at imposing the sanction of disenfranchisement as a further punishment for serious crime. Moral education also appears to be a rationale for this additional sanction. The objective of enhancing civic responsibility through the operation of paragraph 51(e) of the CEA is more elusive. Nevertheless, the Debates of the House of Commons do reveal that some consideration was given to the fact that the impugned provision is capable of sending a message to offenders, and to the general public, about the importance, in a democracy, of the right to vote. [My emphasis.]

[20]      In view of this, the two objectives of the legislation stated by the appellants may be further broken down into two more specific components:

1.         (a) to represent to every citizen that voting is accorded only to those citizens who obey the law; and

(b) to express Parliament’s disapproval of the prisoner’s conduct.[17]

2.         (a) to further punish prisoners who have committed serious offences; and

(b) to encourage the rehabilitation of offenders by providing them with an extra incentive to reform themselves and re-enter society so as to obtain the right to vote.[18]

[21]      The Trial Judge found as a fact that the purposes of the legislation were those two objectives advanced by the appellants.[19] I accept his finding with one observation.

[22]      Because of the history of disenfranchisement of prisoners, I am inclined to believe that paragraph 51(e) of the Act still carries with it some reminiscence of outdated stereotypes concerning the social status of prisoners. Depriving prisoners of the right to vote is an anachronistic concept. This impugned provision is akin to civil death in the law of Edward III, which my colleague Linden J.A. refers to in paragraph 63 of his reasons,[20] and to the old articles 30 to 38 of the Civil Code of Lower Canada which, until 1906, determined that prisoners condemned to death, or to any other corporal punishment for life, be deprived of their civil rights. Changes have now occurred in some jurisdictions. Prisoners in both provincial and federal institutions may vote in provincial general elections in Ontario, Quebec and Newfoundland.[21] In Quebec, all prisoners have been eligible to vote in provincial elections since 1981.[22] British Columbia’s legislation, on the other hand, mirrors the federal provision.[23] Alberta, for its part, is in the process of considering new restrictions in its legislation[24] following the striking down of disqualification provisions by the Alberta Court of Appeal.[25] All federal and provincial prisoners were eligible to vote in the 1992 federal referendum after the decision of the Federal Court of Appeal in Belczowski and the Ontario Court of Appeal in Sauvé No. 1.[26] But, then came the present paragraph 51(e) of the Act.

[23]      The Lortie Commission Report and the House of Commons were careful to note both the criminal and electoral character of the legislation. But, as was observed by the Trial Judge, the objective of enhancing civic responsibilities through the operation of paragraph 51(e) of the Act is more elusive. And the rehabilitation, as we will see later, is nil. The Trial Judge found as a fact that the impugned legislation runs counter to rehabilitation.[27]

[24]      The dominant character of this legislation is unquestionably the further punishment of the detainee following the commitment of serious crimes. This is made clear from the two goals pursued by the legislation, but it is particularly apparent with the breaking down of these goals into the four components mentioned earlier. What the impugned provision does is to express Parliament’s disapproval of the conduct of those who seriously break the law by depriving them of the right to participate in the law-making process. Yet, the link between the crime and the electoral system is somewhat remote. This case is to be distinguished from Harvey v. New Brunswick (Attorney General)[28] where the Supreme Court of Canada considered the constitutionality of a provision suspending the right to vote or sit in the legislature upon conviction for corrupt or illegal practice under the New Brunswick Elections Act [R.S.N.B. 1973, c. E-3]. The disqualification in that case derived from a violation of the electoral process itself. Here, any crime which carries with it detention of two years or more will result in disenfranchisement for those who have the status of detainee at the time of voting so as to enhance the value of the voting process. While both objectives, electoral and penal, are pursued, the penal character of the legislation is the focal point.

[25]      This characterization is essential to an analysis of section 1 of the Charter.

II.         The proper level of scrutiny under the Oakes test

[26]      An infringement of a fundamental right or freedom is only justified under section 1 of the Charter when both components of the Oakes [The Queen v. Oakes, [1986] 1 S.C.R. 103] test are satisfied. The Oakes test itself has been elucidated by later cases. The cardinal principle is, of course, that the test is to be applied flexibly and in context.[29] The first branch of the Oakes test requires a court to determine whether the objectives of the impugned legislation are sufficiently pressing and substantial to justify infringing constitutionally protected rights. The second branch of the Oakes test requires a court to assess whether the means chosen by Parliament to implement the objectives pursued are reasonable and demonstrably justifiable in a free and democratic society. The second branch of this analysis has three elements. First, the measures adopted must be rationally connected to the objectives of the legislation. Second, the measures must impair constitutional rights and freedoms “as little as possible.” The formulation of this second element has varied over the years. The expression “as little as possible” was repeated in Ross v. New Brunswick School District No. 15,[30] but it was also said that “an impairment must be minimal to the extent that it impairs the right no more than is necessary”.[31] In adding this sentence, La Forest J., for the Court, relied on a statement by McLachlin J. in RJR-MacDonald Inc. v. Canada (Attorney General)[32] in which she said that “the government must show that the measures at issue impair the right … as little as reasonably possible” and that “the law must be carefully tailored so that rights are impaired no more than necessary”. A year later, in Eldridge v. British Columbia (Attorney General), the words “no more than is reasonably necessary” were stated.[33] (My emphasis.) But, in the recent case of United Food and Commercial Workers, Local 1518 (U.F.C.W.) v. KMart Canada Ltd.,[34] the Court reverted to the test of RJR-MacDonald Inc. I take the view that these various expressions amount to the same test. Third, there must be proportionality between the deleterious effects of the measures which are responsible for limiting the Charter rights and the objectives of the legislation which have been identified as being of “sufficient importance” and, in case of a partial achievement of the objectives, “there must be a proportionality between the deleterious and salutary effects of the measures”.[35] [Underlining in original.]

[27]      Two other developments of the Oakes test are particularly relevant here. First, the requirement that the objectives of the infringing legislation be “pressing and substantial” has been lowered in the case of civil legislation which can be characterized as “administrative and regulatory” or “social and economic” in character. The rationale for this change appears to have been the difficulty faced by the Government in creating administrative programs tailored to different groups in society in the face of the broad guarantee of equality in section 15 of the Charter. The rationale for the reduced standard was explained by McIntyre J. in Andrews v. Law Society of British Columbia:[36]

In Oakes, it was held that to override a Charter guaranteed right the objective must relate to concerns which are “pressing and substantial” in a free and democratic society. However, given the broad ambit of legislation which must be enacted to cover various aspects of the civil law dealing largely with administrative and regulatory matters and the necessity for the Legislature to make many distinctions between individuals and groups for such purposes, the standard of “pressing and substantial” may be too stringent for application in all cases. To hold otherwise would frequently deny the community-at-large the benefits associated with sound social and economic legislation. In my opinion, in approaching a case such as the one before us, the first question the Court should ask must relate to the nature and the purpose of the enactment, with a view to deciding whether the limitation represents a legitimate exercise of the legislative power for the attainment of a desirable social objective which would warrant overriding constitutionally protected rights.

[28]      The second development of the Oakes test, which must be kept in mind, relates to the second element of the proportionality component of the analysis known as the “minimum impairment” test. In Irwin Toy Ltd. v. Quebec (Attorney General),[37] the Supreme Court of Canada reminded us that “as courts review the results of the legislature’s deliberations, particularly with respect to the protection of vulnerable groups, they must be mindful of the legislature’s representative function”. Where the state behaves as a “singular antagonist” to the individuals whose rights are infringed, it must demonstrate that the means it has chosen to achieve its objective constitute the “least drastic means” available to achieve its goals. This type of state behaviour occurs primarily, but not exclusively, in the criminal law sphere.[38] Alternatively, where state acts are directed towards “the reconciliation of claims of competing individuals or groups or the distribution of scarce governmental resources”,[39] the means chosen by the state will be evaluated according to a less stringent standard. In Stoffman v. Vancouver General Hospital, La Forest J. explained that the rationale for the higher standard in criminal cases is that courts are well suited by their experience in that field to balance the interests of the state against those of individuals, a role they cannot play so well in cases of resource allocation.[40]

[29]      I shall now proceed with an analysis of the first and second branch of the Oakes test.

III.I. First branch of the Oakes test: Are the objectives pressing and substantial enough to justify overriding a protected right?

[30]      The Trial Judge accepted that the objectives of paragraph 51(e) of the Act were pressing and substantial. He approached the matter from a relatively high level of abstraction. He said:[41]

At this stage, attention must be focussed on the democratic ideals which Canada, as a free and democratic society, fosters. There may well be no unified western tradition of political theory, but it is clear from the evidence in this trial that civic and moral responsibility are key components of our liberal democratic traditions. In fact, the preamble to the Charter declares that Canada is founded upon principles that recognize “the rule of law”. The rule of law may be the subject of a number of interpretations, such as a call to law and order, or a legal ordering of social life: J. Rawls, A Theory of Justice (Cambridge, Mass., 1971), at pages 235-243. The ideals of the rule of law express the requirements of legal rules formulated in such a manner as to secure voluntary compliance with the standard of conduct which they set. Of course, while no legal system can expect that all of its laws will be known by the public, it is nevertheless important, as part of the shaping of the voluntary social order, for persons to know in advance what the consequences of their actions might be: E. Colvin, “Criminal Law and the Rule of Law” in Crime, Justice & Codification (Toronto: Carswell, 1986), at page 125.

Paragraph 51(e) of the CEA has a punitive aspect. There is little doubt that retribution is a concept that is not alien to criminal sanctions. Indeed, sentences are invariably partly punitive in nature. As stated by La Forest J. in R. v. Lyons, [1987] 2 S.C.R. 309, at page 329: “In a rational system of sentencing, the respective importance of prevention, deterrence, retribution and rehabilitation will vary according to the nature of the crime and the circumstances of the offender.” See also R. v. Goltz, [1993] 3 S.C.R. 485, at page 503.

Accordingly, I find the objectives of paragraph 51(e) of the CEA to be pressing and substantial.

[31]      In my view, the stringent formulation of the Oakes test applies to the justification of paragraph 51(e) rather than the less stringent one identified in Andrews and Stoffman. Like the provision in question in Sauvé No. 1 and Belczowski, the new version of paragraph 51(e) involves the state interfering in the fundamental rights of a historically vulnerable group of people for a punitive purpose on behalf of society at large. It is a case where the state behaves as a “singular antagonist”. While paragraph 51(e) is not purely penal in character since it attaches civil consequences to incarceration by depriving a detainee of his right to vote, so as to enhance voting rights, such civil consequence involves many of the elements of punishment, and should be subject to the same justificatory standard. Certainly, the impugned legislation cannot be characterized as “administrative and regulatory” or as “social and economic” in nature, so as to trigger the application of the reduced threshold of justification identified by McIntyre J. in Andrews.

[32]      The justifications advanced by the appellants with regard to the voting process are highly abstract. The testimony of the appellants’ experts only establishes that there might be some possible benefits from the legislation over the long term. They certainly did not establish that the legislation was curing any established problems or that its absence would result in any tangible harm.[42] While Parliament is certainly entitled to enact legislation with a view to realizing future benefits or preventing future harms, the benefits and harms must be reasonably established, in the present, before Parliament can rely upon them as its rationale for infringing constitutionally protected rights. While ensuring voluntary compliance with the law is an important goal in democracy, no evidence was presented to establish that the Criminal Code [R.S.C., 1985, c. C-46] and the other rules aimed at accomplishing that objective need to be supplemented by this legislation.

[33]      In light of the abstract and uncertain benefits to be obtained from paragraph 51(e) of the Act, it is with some reservation that I accept the Trial Judge’s finding that the objectives presented by the impugned legislation were in answer to a pressing and substantial concern.

III.II. Second branch of the Oakes test: Are the means chosen reasonable and demonstrably justifiable in a free and democratic society?

III.II.I.   Part one: Rational connection

[34]      The Trial Judge concluded that the objective of enhancing the criminal sanction was the real connecting factor between the means chosen and the goals pursued by the legislation. He was very clear on this when he wrote:[43]

While there is no empirical evidence to suggest that the disenfranchisement of prisoners reduces crime (either generally or specifically), or serves a morally educative function, or could operate as an effective punitive sanction, I find that a rational connection exists between the impugned provision and the stated objective of enhancing the criminal sanction. As an aid to punishment, the provision clearly imposes a sanction, and denounces bad conduct. In the present case, the sanction takes the form of a disenfranchisement, in addition to the loss of liberty. A fundamental democratic right has been removed for crimes committed, and its removal is clearly felt as a deprivation by Mr. Sauvé and Mr. Spence. It is also reasonable to conclude that a morally educative message is sent to offenders, and possibly to the general population, by the imposition of a sanction. Of course, the message may not be heard or understood, but that does not diminish the connection between the means and the second objective. [My emphasis.]

[35]      The Trial Judge accepted the proposition that the disenfranchisement of prisoners sends a morally educative message to offenders and to the general public. But, he noted the absence of empirical evidence which would suggest that the disenfranchisement of prisoners actually serves a morally educative function. In his analysis on minimal impairment, he further elaborated on this and concluded:[44]

Clearly, corrections policy emphasizes the rehabilitation of the offender, and his reintegration into the community. While the defendants argue that disenfranchisement can contribute to rehabilitation, I prefer the plaintiffs’ evidence which suggests that paragraph 51(e) of the CEA hinders the rehabilitation of offenders and their successful reintegration into the community. The provision only serves to further alienate prisoners from the community to which they must return, and in which their families live. Accordingly, the retributive effects of paragraph 51(e) are deleterious in that they are contrary to the purpose and principles contained in the CCRA. [My emphasis.]

[36]      The absence of a rational connection between disenfranchisement and serving the objectives of rehabilitation cannot be overlooked. The enhancement of the criminal law and the sending of a morally educative message are the only objectives rationally connected to the goals pursued by the legislation.

III.II.II.  Part two: Minimal impairment

[37]      The Trial Judge was very conscious of the fact that this was the second time prisoners’ voting rights had been debated before Parliament, and that Parliament is entitled to some margin of appreciation in the choice of measures aimed at the pursuit of legislative goals. What he did not accept in the impugned legislation was its automatic application. He said the following:[45]

The plaintiffs argue that paragraph 51(e) of the CEA fails the minimal impairment test because it is arbitrary in its application. It flows, it is contended, from the sentence rather than from the facts and circumstances which give rise to a particular criminal offence. It is further submitted that only two types of disqualification could survive the minimal impairment test: case-by-case disqualification at the time of sentencing, or a disqualification based on conviction for treason or high treason, as set out in sections 46-48 of the Criminal Code [R.S.C., 1985, c. C-46].

Similarly, the plaintiffs contend that the impugned provision fails the minimal impairment test because there are less intrusive means by which the government could attain its stated objectives. These include the denial of the right to vote at the discretion of the sentencing judge; the Lortie Commission’s recommendation which targeted the most serious offences (those punishable by life imprisonment) and the most serious sentences (those punishable by ten years or more); an offence-oriented approach which would define the specific types of crimes which could be seen as bearing a rational connection to the franchise; and a law which allows for restoration of the right to vote as a result of good behaviour while in prison.

…

As I indicated above, the Lortie Commission recommended that disenfranchisement occur where an offender has been convicted of a crime for which he could potentially receive a punishment of life imprisonment, and for which he did receive a sentence of ten years or more in prison. It is clear that Parliament considered this recommendation and rejected it. Parliament also considered a five-year cutoff and a seven-year cutoff, and rejected both. Furthermore, Parliament considered the voting rights of prisoners in other democratic countries. Eventually, Parliament settled on a two-year cutoff for disqualification. According to the defendants, this limitation allows Parliament to ensure that only those who have committed serious crimes are denied the right to vote.

In this context, the statistics filed by Dr. Colin Meredith demonstrate that, of 654 inmates sampled, who represented 4.5% of the total federal inmate population, there was an average number of 29.5 convictions per inmate. Moreover, 75.3% of the inmates sampled had more than ten convictions during their criminal careers. The Meredith analysis did not find any statistically significant difference between Aboriginal and non-Aboriginal inmates. The statistics do verify that the federal inmate population consists of individuals with long histories of involvement in serious criminal activities. Thus, these statistics appear to support Parliament’s choice for selecting two years as the cutoff for the disqualification of individual offenders who have exhibited bad criminal conduct.

Parliament must have some latitude to choose alternatives. It is for this reason that I reject the plaintiffs’ argument regarding the other alternatives. There is one exception. In this respect, it must be recognized that a final option was available to Parliament: the disenfranchisement of each offender could be imposed on a case-by-case basis, by the sentencing judge. In this way, the disenfranchisement would not be automatic; rather, the right to vote would only be removed by a judge who, as part of the sentencing process, has determined that, in the personal circumstances of the accused, disenfranchisement should occur: R. v. Goltz, supra.

…

The defendants’ experts have provided a number of thoughtful and compelling arguments supporting disenfranchisement. If Parliament decides to pass another law, there is no reason why these arguments could not inform the criteria that could be selected. A sentencing judge could take into account the nature of the crime and the personal circumstances of the accused in conjunction with the principles of sentencing: R. v. Goltz, supra. This process would, in my opinion, be a significantly less intrusive, and equally effective means of infringing a citizen’s democratic right to vote. If a judge is entrusted with the responsibility of taking away a person’s liberty, should he or she not also be charged with the responsibility of determining if disenfranchisement is warranted?

Indecency exists in society generally, and is not only found in correctional institutions. The law as it now stands cannot distinguish the type of offender whose indecency is so profound as to threaten the principles of our free and democratic society. As such, I find that paragraph 51(e) of the CEA fails the minimal impairment component of the Oakes test. [My emphasis.]

[38]      For the same reasons, I believe the legislation in question fails the minimal impairment test.

[39]      It is simply not tailored to the goals pursued by the legislation. This impugned legislation remains a general form of ostracism—or banishment even with a marker of two years or more in prison. To meet the legislative goals, the loss of one’s voting right would be more appropriate if it were gauged on a case-by-case basis. The sentencing judge would be better suited to pronounce disenfranchisement as a further punishment since he would be in a better situation to identify those unfit as democratic voters and deserving of such an additional punishment. Hence, the sentencing judge would deliver Parliament’s message in cases where it would be pertinent to do so. Automatic disenfranchisement is excessive in character. Admittedly, the sentencing judge might apply the sanction sparingly, but the process would be less arbitrary and more significant. While the impugned legislation is narrower in scope than the one in Sauvé No. 1, and admitting that Parliament must have a choice of reasonable alternatives, the Trial Judge felt those alternatives were not acceptable if there existed one way in which the objectives of the impugned provision could truly be served without casting the net too wide. In my view, he correctly applied the minimum impairment test whether encapsulated under the words “as little as possible”, “no more than is necessary”, “no more than is reasonably necessary”, or “as little as reasonably possible”,[46] since I hold that these various expressions amount to the same test. He said, and I repeat:[47]

The law as it now stands cannot distinguish the type of offender whose indecency is so profound as to threaten the principles of our free and democratic society.

[40]      In my view, the Trial Judge was right in concluding as he did. The Trial Judge’s preoccupation on minimum impairment mirrors the concern identified in general terms by the Supreme Court of Canada in the earlier legislation contemplated in Sauvé No. 1. The minimum impairment test has been developed by the case law precisely to meet situations of this type. This is not a situation where the courts are substituting their views for that of Parliament. It is a case where the Charter plays a role in framing the legislation.

III.II.III. Part three: Proportionality

[41]      The proportionality component of the Oakes test requires an examination and a balancing of the deleterious and salutary effects of the legislation under review. In a situation like the present one where the impugned provision does not come close to fully achieving its objectives, there must be proportionality between the deleterious effects of the measures which are responsible for infringing the Charter right and the objectives of the legislation, and there must be proportionality between the deleterious and salutary effects of those measures.[48]

[42]      The Trial Judge found no evidence that the legislation in question actually produces any salutary effects. The Crown had been unable to adduce any evidence of tangible harm flowing from prisoner voting. Most of the evidence offered by the appellants to support the impugned legislation amounts to an educated guess rather than the clear demonstration required by section 1 of the Charter.

[43]      For instance, Dr. Ernest Van den Haag, a criminologist who testified for the Crown, said the following in re-examination:[49]

Q. Doctor, my friend, Mr. Peltz, who examined you earlier, in his cross-examination, he suggested to you that this law had a symbolic effect and you said yes, it did. Is it your view that this law has a solely symbolic effect?

A. No, sir, it does not have a solely … it is conceivable, it seems to me, for instance, that prisoners in prison for violating a particular law might, by voting against the punishment that law imposes, serve their own interests without serving the interests of the population as a whole. So in this sense, it could have possibly a material effect. Also let me clarify, by symbolic, I don’t mean unimportant. I mean quite on the contrary, that, as I pointed out before, it squarely says in effect about prisoners who have seriously violated the law that they are, temporarily at least, excluded from the community, prevented from participating in its affairs, and that makes a moral point as forcefully as it can be made.

[44]      The example given by the witness about prisoners who, if permitted to vote on a penal statute, might put their interest first and not the public good, says nothing about other voters who, for instance, on issues such as free trade or the delivery of medical services (hospital reorganization), might do the same. Moreover, Dr. Van den Haag’s example was, as he stated himself, a possible effect of the legislation, not a proven one.

[45]      Professor Christopher P. Manfredi, who also testified for the Crown, wrote the following in his report:[50]

The inmate voting disqualification, which denies the right to vote to individuals who have manifestly demonstrated their lack of good citizenship, can thus be understood as one means by which a liberal democratic political regime reinforces the notion that voting is, indeed, a mark of good citizenship. In this way, the disqualification contributes to the cultivation of the sense of duty that mitigates the natural incentive of individuals not to vote. The inmate disqualification thus provides an important collective benefit by altering the individual cost-benefit calculation that makes voting irrational and threatens the viability of liberal democracy.

[46]      In his oral testimony, Professor Manfredi could not identify any immediate, tangible harm which Parliament was preventing by means of the impugned provision. He spoke of long-term societal effects. Dr. Manfredi described the “key harm”, which Parliament was preventing, as a “gradual erosion of the link between political participation and civic responsibility”.[51] He said these phenomena could be measured, but admitted that he knew of no evidence to support his claim that harm exists.[52] He conceded he had no evidence of negative effects[53] despite the fact that prisoners do vote fully in some provincial elections and in a number of other countries. His assertions remained theoretical.

[47]      Professor Jean Hampton, a philosopher, summarized well the retributive aspect of the legislation when she stated:[54]

Suspending a criminal’s right to vote is a way for the community that is hurt by a significant crime, such as a murder or a rape or a crime of battery, to say to the criminal….

… you have betrayed us, and you have flouted the values that make this society possible; hence, we deprive you of your right to participate in elections, that you may reflect on how a commitment to these values is what makes a democratic society, and the elective process itself, possible.

Making this communication to the criminal is part of the community’s retributive response to his actions: it is a way for them to vindicate the moral importance of their community, and of the values which they take to be vital to the community’s functioning.

[48]      But, when it came to establish the benefit of this legislation, she stated:[55]

Those who would challenge 51(e) might regard it as kind of mean-minded response to criminals, that can only hurt them, and can do them no good. That perspective misses the way in which effective punishment holds out the hope of morally educating the criminal and the larger society. This law strikes me as potentially highly effective as an educative response. [My emphasis.]

[49]      The benefit, being “potentially highly effective”, remains an educated guess.

[50]      On the whole, the statements made by the appellants’ witnesses are not the kind of evidence to support a section 1 analysis. In RJR-MacDonald Inc. v. Canada (Attorney General),[56] McLachlin J. described the section 1 analysis in the following terms:

The process is not one of mere intuition, nor is it one of deference to Parliament’s choice. It is a process of demonstration. This reinforces the notion inherent in the word “reasonable” of rational inference from evidence or established truths. [Emphasis in original.]

[51]      There was no real demonstration in the case at bar. The Trial Judge concluded, rightly in my view, that:[57]

… the salutary effects upon which the defendants rely are tenuous in the face of the denial of the democratic right to vote, and are insufficient to meet the civil standard ….

[52]      Parliament, when it voted on paragraph 51(e) of the Act, had before it the research paper of P. Landreville and L. Lemonde’s of the Lortie Commission.[58] It chose not to follow the recommendations of these authors. What these authors said, however, expresses well the values embodied in the Charter. The following is worth reading:[59]

Refusing to grant the right to vote to inmates must be seen as a remnant of the exclusion practices of the past.

On the one hand, these practices involved excluding criminals from society through banishment, deportation, loss of citizenship or civil death. When criminals were excluded, they lost their rights. On the other hand, universal suffrage has progressed very gradually, even in the most democratic countries. For a long time, the right to vote was restricted to specific groups of citizens. The poor, the illiterate, Blacks, Aboriginal people and women were all excluded from the democratic process. People who were convicted or incarcerated were excluded both from society and from the right to vote.

The modern trend is toward equality of rights and participation in political life by everyone, even the members of unpopular or marginal groups. Canada may be seen as one of the front runners among democratic countries when it comes to legal and political equality. As we have seen earlier, Canadian courts and correctional policies recognize that inmates retain all of their civil rights other than those expressly denied them by law. The Canadian Charter of Rights and Freedoms spells out the equality of rights for all Canadians, including the right to vote. Exclusions are becoming less and less acceptable, and we believe that there ought to be no exceptions to the right to vote and that the Canada Elections Act should be revised to grant the right to vote to all prison inmates. [My emphasis.]

[53]      What this case decides today is whether this last remnant of exclusionary practices can resist the Charter. In my view, the impugned legislation cannot and is therefore invalid.

IV.       Conclusion on the appeal

[54]      I would dismiss this appeal with costs.

B.        The cross-appeal

[55]      I agree with my colleague Linden J.A. that section 15 of the Charter is not infringed and that the cross-appeal should be dismissed with costs.

* * *

The following are the reasons for judgment rendered in English by

Linden J.A.:

I.          Introduction

[56]      This case is another episode in the continuing dialogue between courts and legislatures on the issue of prisoner voting.[60] In 1992 and 1993, two appeal courts and the Supreme Court of Canada held that a blanket disqualification of prisoners from voting, contained in earlier legislation which was challenged, violated section 3 of the Charter and could not be saved by section 1 of the Charter.[61] Parliament responded to this judicial advice by enacting legislation aimed at accomplishing part of its objectives while complying with the Charter. That legislation, which is being challenged in this case, disqualifies from voting only prisoners who are serving sentences of two years or more.

[57]      The main issue in this appeal is whether paragraph 51(e) of the Canada Elections Act (the CEA or the Act) can be saved by section 1 of the Canadian Charter of Rights and Freedoms (the Charter). More particularly, the issue in this appeal is whether a provision which prohibits “every person who is imprisoned in a correctional institution serving a sentence of two years or more” from voting meets the minimal impairment and proportionality tests mandated by section 1 of the Charter. In thoughtful and lengthy reasons, the Trial Judge found that it did not.[62] For the reasons which follow, I believe that it does.

[58]      The respondents in this case also argue that subsection 15(1) of the Charter is infringed by this provision. I am however, of the view that subsection 15(1) of the Charter is not infringed by this legislation.

[59]      The respondent, Sauvé, was, among other people, involved in the shooting death of Bill Matiyek on October 18, 1978. Mr. Sauvé, at the time a member of the Satan’s Choice Motorcycle Club, was convicted of first degree murder and sentenced to 25 years’ incarceration. Mr. Sauvé filed a statement of claim on September 24, 1993, seeking a declaration that paragraph 51(e) of the CEA is contrary to sections 3 and 15 of the Charter. Mr. Sauvé was released on parole in May, 1994. The respondents McCorrister et al., inmates at the Stony Mountain Institution in Manitoba, filed a statement of claim in May, 1994, seeking the same relief as the respondent Sauvé. Three of the McCorrister respondents, Sheldon McCorrister, Lloyd Knezacek, and Aaron Spence, are members of the Native Brotherhood Organization, a group which represents Aboriginal inmates at the Stony Mountain Institution. The two actions were joined and heard together by the Trial Judge in June 1995. On December 27, 1995, the Trial Judge gave judgment, striking down paragraph 51(e) of the CEA, holding that the (conceded) violation of section 3 of the Charter could not be saved by section 1 of the Charter. The Crown now appeals from that judgment.

[60]      Whether paragraph 51(e) of the Act is good penal policy or good public policy is not at issue in this appeal. It is not the role of this Court to decide what works with regard to penal policy and what does not. It is not the role of this Court to determine what theories of penology should be adopted by our elected legislatures. This case is about what, if anything, Parliament may or may not do to interfere with prisoner voting rights within the bounds of section 1 of the Charter. At issue is whether the statutory prohibition is sufficiently tailored and appropriately proportional, or whether Parliament must try again to fashion a still narrower bar, adopt a different approach, or abandon the objective altogether. The matter is both important and complex, with no easy answer readily apparent.

II.         Crown concession and relevant legislative provisions

[61]      The Crown conceded before this Court that paragraph 51(e) of the CEA violates section 3 of the Charter, but argues that the provision is saved by section 1 of the Charter.

[62]      The relevant statutory provisions are paragraph 51(e) of the CEA, section 3 of the Charter, and section 1 of the Charter. Paragraph 51(e) of the CEA reads as follows:

51. The following persons are not qualified to vote at an election and shall not vote at an election:

…

(e) every person who is imprisoned in a correctional institution serving a sentence of two years or more.

Section 3 of the Charter specifies:

3. Every citizen of Canada has the right to vote in an election of members of the House of Commons or of a legislative assembly and to be qualified for membership therein.

Section 1 of the Charter provides that:

1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

III.        A brief history of prisoner disenfranchisement in pre-Charter Canada

[63]      Prisoner disenfranchisement can be traced back to the notion of “civil death”, by which one consequence of being convicted of a felony was the loss of all civil rights. One of the earliest written records of civil death occurs in the law of Edward III, De Catallis Felonum.[63] This statute laid down the doctrine of forfeiture and stated that an outlaw forfeited not only his personal property but every possible right and means of acquiring property.

[64]      The Constitutional Act, 1791,[64] which established Upper and Lower Canada, specifically provided for prisoner disenfranchisement. It stated, in part, that “no Person shall be capable of voting at any Election of a Member to serve in such Assembly, in either of the said provinces … who shall have been attained for Treason or Felony in any Court of Law within any of His Majesty’s Dominions”.[65] Subsequently, section 41 of the Constitution Act, 1867 [30 & 31 Vict., c. 3 (U.K.) (as am. by Canada Act 1982, 1982, c. 11 (U.K.), Schedule to the Constitution Act, 1982, Item 1) [R.S.C., 1985, Appendix II, No. 5]] preserved the status quo and authorized Parliament to establish the qualifications for membership in the House of Commons and for voting. That section reads as follows:

41. Until the Parliament of Canada otherwise provides, all Laws in force in the several Provinces at the Union relative to the following Matters or any of them, namely,—the Qualifications and Disqualifications of Persons to be elected or to sit or vote as Members of the House of Assembly or Legislative Assembly in the several Provinces, the Voters at Elections of such Members, the Oaths to be taken by Voters, the Returning Officers, their Powers and Duties, the Proceedings at Elections, the Periods during which Elections may be continued, the Trial of controverted Elections, and Proceedings incident thereto, the vacating of Seats of Members, and the Execution of new Writs in case of Seats vacated otherwise than by Dissolution,—shall respectively apply to Elections of Members to serve in the House of Commons for the same several Provinces.

Provided that, until the Parliament of Canada otherwise provides, at any Election for a Member of the House of Commons for the District of Algoma, in addition to Persons qualified by the Law of the Province of Canada to vote, every Male British Subject, aged Twenty-one Years or upwards, being a Householder, shall have a Vote.

[65]      Canada’s first electoral law, The Electoral Franchise Act,[66] made no specific reference to prisoner disenfranchisement. Subsection 3(1) of that Act, however, required that voters be “of full age of twenty-one years, and … not by this Act or any law of the Dominion of Canada, disqualified or prevented from voting.” Thus, the voter disqualification contained in the Constitutional Act, 1791 and preserved by section 41 of the Constitution Act 1867 was likely the law in force at that time.

[66]      The matter was clarified in 1898. In that year, The Franchise Act, 1898 denied the vote in federal elections to “[a]ny person, who, at the time of an election, is a prisoner in a jail or prison undergoing punishment for a criminal offence”.[67] This blanket prohibition of prisoners is nearly identical to that which was in force at the time of the enactment of the Charter, and which was successfully challenged in earlier cases. The former, unconstitutional provision read as follows:

14. …

(4) The following persons are not qualified to vote at an election, and shall not vote at an election:

…

(e) every person undergoing punishment as an inmate in any penal institution for the commission of any offence;[68]

[67]      The current law is much less intrusive, denying the vote only to those serving two years or more and not to all of those in penal institutions.

IV.       Prisoner disenfranchisement in the post-Charter era

[68]      This issue has been dealt with before in various courts including this Court. There have been a number of cases challenging this section or its immediate predecessor invoking section 3 of the Charter. These cases will be analysed first in order to place this appeal in context. There have also been several cases challenging similar provincial statutes, which I will not summarize here.[69] It will be seen that the cases considering the earlier, blanket prohibition on prisoner voting evince little unanimity of reasoning, as might be expected in cases dealing with difficult, politically charged matters.

[69]      In 1983, Taylor J. of the British Columbia Supreme Court upheld the validity of the former prisoner disenfranchisement provision.[70] His Lordship reasoned that, since prisoners could not make a “free and democratic electoral choice,” it was appropriate for Parliament to deny them the vote:

Imprisonment, as a punishment for breach of the criminal law, is clearly justifiable in a free and democratic society. It follows that denial to prisoners of those constitutional rights which, of necessity, cannot be exercised by persons serving a sentence of imprisonment is also justifiable and must be taken to be authorized by s. 1 of the Charter.

[70]      One year later, in Attorney General of Canada v. Gould,[71] this Court considered whether to issue an interlocutory motion for an injunction enabling the applicant to vote in an upcoming election. While all of the members of this Court agreed that the provision as it was then written constituted a prima facie violation of the Charter, they disagreed on whether an injunction should issue enabling the applicant to vote. No guidance was given in those cases regarding any section 1 analysis of provisions limiting inmate voting rights. A similar issue arose in the later case of Lévesque v. Canada (Attorney General), where Rouleau J. held that an injunction should be issued permitting the applicant, a prisoner, to vote in an upcoming election.[72]

[71]      In 1988, Hirschfield J. of the Manitoba Court of Queen’s Bench came to the same conclusion as Scollin J., who had held inoperative provisions of the Manitoba Elections Act [C.C.S.M., c. E-30] in an earlier case.[73] In coming to the same conclusion, Hirschfield J. quoted extensively from the earlier judgment.[74] In notable dicta, however, Hirschfield J. opined that, while excluding all prisoners from voting was too broad, the disqualification of federal prisoners serving sentences for indictable offences would likely be acceptable:

Had the words “penal institution” been defined to mean only a federal penitentiary, and had “any offence” been defined to mean an indictable offence, the result which I am about to announce would have been radically different. In my view the proportionality test referred to in Oakes would then have been weighted in favour of disqualification.[75]

[72]      The Manitoba Court of Appeal reversed the decision of Hirschfield J.,[76] holding that the prisoner disqualification was a reasonable limit on the right to vote. After summarizing voter disqualification provisions in other democratic societies, Monnin C.J.M. wrote that the connection between voter disqualification and standing as a candidate justified the complete disqualification of prisoners:

I conclude that Hirschfield J. erred when he ruled that the breach of s. 3 was not saved by s. 1 of the Charter. Sec. 14(4)(e) is a reasonable and demonstrably justified limit on the right to vote which is guaranteed to Canadian citizens by s. 3 of the Charter and it is therefore valid legislation. If inmates are enfranchised they will automatically be given the right to stand as candidates for federal elections. There is no disqualification provision in the Act with respect to a person’s right to offer his/her services as a candidate. One only has to be a voter who can enter his name on the enumeration list and meet the residential qualification. One can recall that in recent years a member of the Sinn Fein was elected to the Mother of Parliaments at Westminster while in a penitentiary in Northern Ireland. I do not recall whether his election was annulled by the High Court of Justice of the British Parliament which has the right to rule on the eligibility of its membership.

…

The right to vote is therefore not an absolute one although it is essential in a democracy. Unfortunately, at times, for valid reasons, that right cannot be enforced.[77]

[73]      Three days after the decision of Hirschfield J., Van Camp J. of the Ontario Court upheld as constitutional the same provision which Justice Hirschfield had held to be of no force and effect. Van Camp J. noted the historic value placed on ensuring a responsible electorate:

The history of the disqualifications of voters over the years has reflected the different understanding of who would be responsible. The specific question now before the court is whether it is justifiable that the person who breaks the law should participate in the choice of those who make the law. The right of every citizen to vote has not been a part of our history. Historically, a vote has been deemed to more likely be responsible if the person casting it:

(1) had a demonstrable stake in the community and its public affairs;

(2) took an active interest in public affairs, and

(3) was adequately informed about public issues.

Consequently, there have, in the past, been requisite property, wealth, literacy and residential qualifications. However, these different qualifications have been repealed as such criteria have come to be seen as inadequate tests of the desired responsibility. There remains then the citizenship qualification with its included qualification of age. Until now, there has persisted the disqualification of criminals and those involved in corrupt electoral practices on the ground that they are not responsible citizens, and that they have demonstrated beyond all doubt their lack of commitment to the well-being of the community.[78]

[74]      While Van Camp J. reasoned that neither punishment nor administrative convenience would be sufficient to justify the disqualification of all prisoners, she accepted that the disqualification could stand as a proxy for a responsible electorate:

I would accept the submission that the objective of punishment is not of such importance that it should override the basic right.

I would agree with the opinion of Reed J. in Gould v. A.-G. Can., … and of Rouleau J. in Lévesque v. A.-G. Can. …, that administrative or security reasons cannot prevent the exercise of a constitutionally recognized right and that imprisonment does not necessarily entail as a consequence the loss of voting rights.

However, it seems to me that Parliament was justified in limiting the right to vote with the objective that a liberal democratic regime requires a decent and responsible citizenry. Such a regime requires that the citizens obey voluntarily; the practical efficacy of laws relies on the willing acquiescence of those subject to them. The state has a role in preserving itself by the symbolic exclusion of criminals from the right to vote for the lawmakers. So also, the exclusion of the criminal from the right to vote reinforces the concept of a decent responsible citizenry essential for a liberal democracy.[79] [Citations omitted; emphasis added.]

[75]      With regard to proportionality, Van Camp J. held that the temporary disenfranchisement of all prisoners was proportionate. She wrote that:

The objective of the Act was the protection of an essential value to our present democratic society, in the recognition of the importance of the vote. The statute was not a denial of the right. The statute does not discriminate on the basis of any inborn characteristics of race or sex. The right itself was not a confirmation of a long-standing pre-existing right. The Charter provision itself asked for limitation. The criteria of limitation provided for by the representatives of the people were reasonable in light of the history of the right to vote, the effect of the right to vote, the practice of other free and democratic societies. Parliament has carefully considered the extent to which those convicted should be disenfranchised. It has not removed the right of citizenship. It has not removed the right from all those who have been convicted. It has not provided that the disqualification is for all time. It has not made the return of the right dependant upon any subsequent decision. The return is automatic as soon as the person ceases to be an inmate. It has provided for the return of the vote before the sentence is completed, as soon as the prisoner has shown the requisite for a gradual return to society. The disqualification is in fact upon those who have chosen to disqualify themselves. It is not the loss of any of the fundamental freedoms.[80] [Emphasis added.]

[76]      In 1991, before the appeal from the decision of Van Camp J. was heard, Strayer J. (as he then was) considered the same provision in the case of Belczowski v. Canada.[81] Strayer J. rejected the objectives of “affirming and maintaining the sanctity of the franchise in our democracy” and “preserving the integrity of the voting process” as being pressing and substantial objectives. Neither of these specific objectives, however, were argued before this Court. A third objective for the former provision was presented to Strayer J., that being to sanction offenders. Strayer J. considered this objective to be rationally connected to the legislation, but held that the legislation was neither minimally impairing nor proportionate. He wrote that:

Again, however, it cannot be said that this means of punishment impairs the section 3 right “as little as possible”. Instead it directly and completely abolishes that right for the period of imprisonment. In this it is in contrast to incidental abridgment, brought about by imprisonment, of other Charter rights and freedoms such as freedom of association or assembly or expression.

Finally, with respect to voting disqualification as punishment, the government has not demonstrated to my satisfaction that the outright denial of the vote of every prison inmate is proportional to this objective. First, it may be noted that paragraph 51(e) applies no matter what the seriousness of the crime may be for which the inmate is being punished. Secondly, the actual effect on the inmate’s right to vote will be quite arbitrary, depending on fortuitous circumstances such as the timing of federal elections in relation to the period he happens to serve his sentence. Thus someone in prison for two weeks for non-payment of parking fines could lose his vote for four years because his sentence happened to coincide with a federal election. On the other hand, someone sentenced to prison for five years for fraud or sexual assault and released on parole after three and one-half years might never miss the opportunity to vote. Thus there is no necessary coordination between serving of a prison sentence and the actual loss of a right to vote. Thirdly, there is a lack of proportionality between the objective and the denial of the vote in that corrections theory in Canada for the last fifty years has moved in the direction of rehabilitation and the preparation of inmates for reentry into society.[82]

[77]      The decision of Strayer J. was upheld by this Court. Speaking for the Court, Hugessen J.A. found that the objectives as presented were too abstract to warrant the infringement of constitutionally protected rights. He wrote that:

Viewed together and collectively, the most striking point about the alleged objectives of paragraph 51(e) is that they are all symbolic and abstract. The appellant admits as much, but maintains that this fact does not prevent them from being legitimate objectives for legislation. With respect, it seems to me that this misses the mark. It is, of course, true that legislation may legitimately have a purely symbolic objective. The question on the first branch of the Oakes test, however, is not the legitimacy of the legislative purpose but rather its importance, that is to say whether it is “pressing and substantial”. For my part, I must say that I have very serious doubts whether a wholly symbolic objective can ever be sufficiently important to justify the taking away of rights which are themselves so important and fundamental as to have been enshrined in our Constitution. To accept symbolism as a legitimate reason for the denial of Charter rights seems to me to be a course fraught with danger. Even on the lower test of a “desirable social objective” suggested in Andrews, I would have thought that such objective would have to translate into some real intended benefit and not merely some abstract or symbolic notion. To adopt the other course would, it seems to me, expose us to Voltaire’s famous jibe that the English had executed Admiral Byng on his own quarterdeck “pour encourager les autres”.

Assuming, however, for the sake of argument, that a purely symbolic objective may be sufficiently serious in some circumstances, it is my view that it cannot be so in this case. Depriving prisoners of the vote is not a ringing and unambiguous public declaration of principle. On the contrary it is an almost invisible infringement of the rights of a group of persons who, as long as they remain inside the walls are, to our national disgrace, almost universally unseen and unthought of. If, as I think, therefore, the alleged symbolic objective is one whose symbolism is lost on the great majority of citizens, it is impossible to characterize that objective as pressing or substantial.[83]

[78]      Hugessen J.A. refused to accept the government’s objectives. He opined that the true objective of prisoner disqualification was to further degrade the inmate:

Alternatively, and far less commendably, it would appear to me that the true objective of paragraph 51(e) may be to satisfy a widely held stereotype of the prisoner as a no-good almost sub-human form of life to which all rights should be indiscriminately denied. That, it need hardly be said, is not an objective which would satisfy section 1 of the Charter.[84]

[79]      Hugessen J.A. did, however, suggest that a denial of the right to vote for people convicted of certain crimes might be acceptable. To him it was unacceptable that all prisoners be disenfranchised as a consequence of their condition. He wrote that:

A denial of the right to vote for persons convicted of treason or felony can readily be understood as a punishment for those crimes. A similar denial imposed only on those who are actually in prison looks more like a consequence of that condition than a sanction for the conduct which brought it about in the first place.[85]

[80]      With regard to proportionality, Hugessen J.A. agreed with Strayer J. that the former provision was not proportional. Hugessen J.A. held that the legislation failed at every stage of the proportionality test. First, he contended that the fact of being in prison was not a rational indication of irresponsible citizenship:

First, there is the requirement that paragraph 51(e) be rationally connected to the alleged objectives. It is not. The fact of being in prison is not, by any means, a sure or rational indication that the prisoner is not a decent and responsible citizen. I have already mentioned fine defaulters who shockingly constituted a huge proportion of our prison population. By no means can they be described as ipso facto indecent and irresponsible. It is also not impossible in our society for persons to be in prison for reasons of conscience and I doubt that as a society we feel that such persons are not decent and responsible whatever else we might think of them.[86]

[81]      With regard to minimal impairment, Hugessen J.A. held that by virtue of the design of the legislation, the right to vote was taken away in an irregular and irrational way:

There is little that need to be said of the second branch of this part of the test which requires that the legislative measure impair the guaranteed right as little as possible. I would only note that, not only is the right taken away altogether, but, because of the very nature of the right to vote itself, it is taken away in an irregular and irrational pattern: persons who happen to be in prison on enumeration day, or voting day, no matter how short their sentence, lose the right to vote; others may serve up to four years and three hundred and sixty-four days in prison and never be deprived of the franchise at all.[87]

[82]      With regard to proportionality, Hugessen J.A. held that the legislation was both over- and under-inclusive in that it made no attempt to balance the seriousness of the offender’s conduct to the effects of the legislation:

Finally, the third branch of the test requires an examination of the proportionality between the effect of the legislation and its objectives. For reasons which have already been suggested, paragraph 51(e) cannot meet this test. I have already commented on the over- and under-inclusiveness of the legislation when viewed in the light of its alleged objectives. I have also indicated that the legislation makes no attempt to weigh, assess or balance the seriousness of the conduct which may have resulted in imprisonment and the resultant deprivation of a Charter guaranteed right. Finally, I have indicated that as a necessary result of the legislation, and not merely of its imperfect application, its actual operation in any particular case will depend on wholly fortuitous circumstances which bear no relationship either to the alleged objectives or to the conduct of the prisoners whose rights are thus taken away. Even assuming the alleged objectives to be valid, paragraph 51(e) simply cannot be characterized as a measured and proportionate means of achieving them with due regard for the importance of the rights taken away.[88]

[83]      Shortly after this Court’s decision in Belczowski, the Ontario Court of Appeal overturned the judgment of Van Camp J. in the Sauvé case. Speaking for that Court, Arbour J.A. (as she then was) agreed with Hugessen J.A. that the symbolic nature of the objectives detracted from its importance as a justification for the violation of the Charter-protected right to vote. In obiter, which the parties to this case vigorously debated, she wrote that:

I would also add that the slow movement toward universal suffrage in Western democracies took an irreversible step forward in Canada in 1982 by the enactment of s. 3 of the Charter. I doubt that anyone could now be deprived of the vote on the basis, not merely symbolic but actually demonstrated, that he or she was not decent or responsible. By the time the Charter was enacted, exclusions from the franchise were so few in this country that it is fair to assume that we had abandoned the notion that the electorate should be restricted to a “decent and responsible citizenry”, previously defined by attributes such as ownership of land or gender, in favour of a pluralistic electorate which could well include domestic enemies of the state.[89]

[84]      While Arbour J.A. accepted that the most plausible objective presented was to sanction offenders, she felt that objective was missed by a provision which punishes inmates generally. She wrote that:

If the objective of s. 51(e) is to punish offenders, that objective is missed altogether by a provision that punishes inmates and that is therefore both over- and under-inclusive. Whether this is viewed as a question of proportionality or objective, the result remains that it fails as a constitutional justification. The same can be said of all three objectives which, even taken collectively, are either insufficiently important or unacceptable objectives or are expressed in means which are not rationally connected with the objectives or which impair the right to vote far more than necessary. I find no need to analyze in detail the proportionality flaws of this legislation as I agree substantially with the reasons of both Strayer J. and Hugessen J.A. in Belczowski, supra, on the question of proportionality.[90]

[85]      The Supreme Court dismissed appeals from both the decisions of the Ontario Court of Appeal and this Court at the same time. I reproduce in their entirety, the reasons for judgment of the Supreme Court which are as follows:

We are all of the view that these appeals should be dismissed.

The Attorney General of Canada has properly conceded that s. 51(e) of the Canada Elections Act, R.S.C., 1985, c. E-2, contravenes s. 3 of the Canadian Charter of Rights and Freedoms but submits that s. 51(e) is saved under s. 1 of the Charter. We do not agree. In our view, s. 51(e) is drawn too broadly and fails to meet the proportionality test, particularly the minimal impairment component of the test, as expressed in the s. 1 jurisprudence of the Court.

Accordingly, the first constitutional question is answered in the affirmative and the second constitutional question is answered in the negative.

Consequently, both appeals are dismissed with costs.[91]

It should be noted that the Supreme Court says only that the former section was “drawn too broadly” and specifically that it failed the minimal impairment component of the proportionality test laid down by the Court in section 1 jurisprudence. There is nothing in the reasons of the Supreme Court that grants section 3 of the Charter any immunity from potential limitations drawn within the bounds of section l of the Charter.

[86]      Shortly before the Supreme Court’s decision in the Sauvé and Belczowski matters, Parliament passed amendments to the CEA which contained the prisoner disenfranchisement provision now before this Court.[92] This is, therefore, a new law being challenged, one that Parliament has sought to enact in conformity with the Charter and the judicial decisions. Previous cases may consequently be helpful for an understanding of the context of this appeal, but are in no way binding on this Court. The previous provision applied to all prisoners, regardless of their crime or term of incarceration. The current provision is significantly different, being much more limited, and, hence, requires fresh consideration.

V.        Is paragraph 51(e) saved by section 1 of the Charter?

1. Introduction: the proper approach to section 1 of the Charter?

[87]      Any approach to section 1 of the Charter must begin with the words of section 1 itself:

1.  The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

[88]      The Supreme Court has frequently emphasized that, in the section 1 analysis, context matters. The Supreme Court has instructed that proportionality can only be measured through close attention to the details and facts of the case.[93] The Oakes test is therefore not meant to be a rigid test, and indeed it cannot be. Concepts such as minimal impairment and proportionality must be viewed in the light of the case before the Court. Giving such concepts a standardized meaning applicable to each and every case would create a test which would be unhelpful in many cases. As I will discuss below, this is a case in which the application of the Oakes test is particularly sensitive to context.

[89]      It is only in the context of the particular case that the proper standard of proof can be established. In many cases, scientific and conclusive proof of the effects of legislation is impossible. While the onus always remains on the government to justify Charter violations demonstrably on the balance of probabilities, a court should be willing to find causal connection between legislation and its intended benefits on the basis of reason, deduction, or inference.[94]

[90]      The basic framework for a section 1 analysis set out in The Queen v. Oakes, remains in place, but has been elucidated by subsequent decisions. The analytical framework was restated in Egan v. Canada, and quoted with approval in Eldridge and Vriend:

A limitation to a constitutional guarantee will be sustained once two conditions are met. First, the objective of the legislation must be pressing and substantial. Second, the means chosen to obtain this legislative end must be reasonable and demonstrably justifiable in a free and democratic society. In order to satisfy the second requirement, three criteria must be satisfied: (1) the rights violation must be rationally connected to the aim of the legislation; (2) the impugned provision must minimally impair the Charter guarantee; and (3) there must be a proportionality between the effect of the measure and its objective so that the attainment of the legislative goal is not outweighed by the abridgement of the right. In all s. 1 cases the burden of proof is with the government to show on a balance of probabilities that the violation is justifiable.[95]

[91]      The first segment of the Oakes test is well known. The objective of the legislation must be “pressing and substantial” in order to be of sufficient importance to warrant overriding a constitutionally protected right or freedom.

[92]      If this requirement is met, the second requirement involves a proportionality test. The proportionality test is comprised of three components. First, the measure limiting the Charter right must be rationally connected to the intended objective. In other words, the measure must be designed to achieve its objective without being arbitrary, unfair, or based on irrational considerations. Second, the limiting measures must impair the Charter right no more than is necessary.[96] Third, the effects of the measure must be proportional to the significance of the objective which is to be achieved, and the salutary and deleterious effects of the measure must be proportional. A provision limiting a Charter right that fails to satisfy any one of these criteria will not be saved under section 1.

2. Are the objectives of this legislation sufficiently pressing and substantial to warrant compromise of a Charter right?

[93]      The first task required of any judge assessing the justification presented for a Charter infringement is to identify the objectives underlying the legislation in question. The Crown submits that there are two objectives underlying this legislation, namely those accepted by the Trial Judge. Those objectives are as follows:

(a) the enhancement of civic responsibility and respect for the rule of law; and,

(b) the enhancement of the general purposes of the criminal sanction.

In this case, the Trial Judge considered the submissions of the parties, the legislative history of the provision, and the text of the provision, finding that there was evidence that the provision was enacted to advance both of these objectives. He concluded that:

Specifically, the legislative text, in conjunction with the proceedings of the Special Committee on Electoral Reform, reveal that the provision is clearly directed at imposing the sanction of disenfranchisement as a further punishment for serious crime. Moral education also appears to be a rationale for this additional sanction. The objective of enhancing civic responsibility through the operation of paragraph 51(e) of the CEA is more elusive. Nevertheless, the Debates of the House of Commons do reveal that some consideration was given to the fact that the impugned provision is capable of sending a message to offenders, and to the general public, about the importance, in a democracy, of the right to vote.[97]

[94]      Courts of appeal are slow to overturn findings of adjudicative facts, as it is the trial judge who receives the most complete presentation of the facts. Hence, much deference is paid to the findings of fact at trial.

[95]      While findings of legislative fact are not entitled to the same level of deference as findings of adjudicative fact,[98] such findings are still entitled to respect. In this case I have reviewed the evidence, and I agree with the Trial Judge that the government has proved on a balance of probabilities that the objectives it contends for indeed motivated the passage of this legislation. The Trial Judge made no error in accepting the objectives as he did.

[96]      In reviewing the evidence, I have reviewed the Parliamentary debate. I note with interest that Parliament, both in general session and in Committee, debated this measure vigorously. The Parliamentary Committee reviewing the matter carefully considered the submissions of the Royal Commission on Electoral Reform and Party Financing (the Lortie Commission), recommending disenfranchisement to all those prisoners serving sentences of 10 years or more. While in session, Parliament debated and ultimately rejected a motion to repeal the disenfranchisement, and an alternative motion to trigger the disenfranchisement only after a sentence of five years or more is handed down.[99]

[97]      It is noteworthy that the respondents do not challenge the legislative objectives as found. Indeed Mr. Peltz’ very fine written submissions accept the twin objectives of this legislation and attempts to equate them to those objectives presented—and rejected—in the earlier Sauvé matter.[100] While this legislation was motivated by similar objectives which supported the earlier, total disenfranchisement, this legislation differs from the earlier provision enough to demand fresh consideration. In debating this issue, the Lortie Commission Report, the Parliamentary Committee, and the House of Commons were all careful to note the hybrid criminal and electoral nature of this provision. Further, this provision was enacted after full consideration of the Charter; Parliament enacted this new provision in response to the decisions of this Court and the Ontario Court of Appeal, affirmed by the Supreme Court of Canada, which declared the earlier law unconstitutional. Wary that the current provision would also be challenged in the courts, Parliament sought to further objectives which were less “symbolic” and which pursued what Parliament believed to be sound penal and electoral policy.[101]

[98]      Before this Court, the Alberta government[102] argued enthusiastically that the electoral nature of these provisions should be recognized. Specifically, the Government of Alberta is concerned that if this Court characterizes this legislation purely as an exercise of the criminal law power, then the ability of provincial governments to disenfranchise prisoners will be called into question. This Court is not prepared in this case to consider the extent to which provincial governments may disenfranchise prisoners in provincial elections. In my opinion, the federal government enacted this legislation both as an exercise of its criminal law power and as an exercise of its right to legislate about electoral law. This is evident from what was said during the Parliamentary debate which preceded its enactment. I would say nothing about the extent to which electoral objectives may or may not properly motivate similar provincial legislation.

[99]      Further, I would leave to philosophers the determination of the “true nature” of the disenfranchisement. It may be argued that this legislation does different things—it imposes a civil consequence, it fixes a civil disability, it imposes a criminal penalty, it furthers a civic goal, it promotes an electoral goal, or it is part of the sentencing process. I believe that these arguments, made alone, are of limited assistance. There are elements of all these ideas and ideals at work here. Nonetheless, we are seized with the question of whether this legislation, in its objectives and effects, can be saved by section 1 of the Charter. In doing so, the words of the legislation and, to a lesser extent, the debates in Parliament are of paramount importance. Focussing exclusively on the nature of the disenfranchisement treats the Parliamentary process as something other than the complex, collective process that it is.

[100]   Having found that the objectives of enhancement of civic responsibility and respect for rule of law and enhancement of the penal sanction were those which motivated Parliament, I agree with the Trial Judge that those objectives are sufficiently pressing and substantial as to warrant an infringement of a Charter right. While Mr. Peltz, on behalf of the McCorrister respondents, argues that these objectives are too abstract and symbolic to warrant the infringement of a Charter right, and while the breadth of the first objective is very wide indeed, I accept that fostering civic responsibility and respect for the rule of law is important enough, in some cases, to warrant compromise of Charter rights.[103] I am also of the view that the penal sanction may be utilized in ways that may compromise Charter rights. The Trial Judge considered the matter and concluded as follows:

At this stage, attention must be focussed on the democratic ideals which Canada, as a free and democratic society, fosters. There may well be no unified western tradition of political theory, but it is clear from the evidence in this trial that civic and moral responsibility are key components of our liberal democratic traditions. In fact, the preamble to the Charter declares that Canada is founded upon principles that recognize “the rule of law”. The rule of law may be the subject of a number of interpretations, such as a call to law and order, or a legal ordering of social life: J. Rawls, A Theory of Justice …, at pages 235-243. The ideals of the rule of law express the requirements of legal rules formulated in such a manner as to secure voluntary compliance with the standard of conduct which they set. Of course, while no legal system can expect that all of its laws will be known by the public, it is nevertheless important, as part of the shaping of the voluntary social order, for persons to know in advance what the consequences of their actions might be: E. Colvin, “Criminal Law and The Rule of Law” in Crime, Justice & Codification …, at page 125.

Paragraph 51(e) of the CEA has a punitive aspect. There is little doubt that retribution is a concept that is not alien to criminal sanctions. Indeed, sentences are invariably partly punitive in nature. As stated by La Forest J. in R. v. Lyons, … at page 329: “In a rational system of sentencing, the respective importance of prevention, deterrence, retribution and rehabilitation will vary according to the nature of the crime and the circumstances of the offender.” See also R. v. Goltz, … at page 503.[104]

[101]   I adopt the conclusion of the Trial Judge on this issue.

3. Are these objectives rationally connected to the law which seeks to further them?

[102]   In order to show that legislation which violates a Charter right is saved by section 1 of the Charter, the government must demonstrate that there is a rational connection between the objectives which it seeks to further and the means by which it has chosen to do so. As I have recently stated,[105] it is my view that a “rational connection” is just that—not a “proven connection” or an “established connection” but a “rational connection”. Discussion of the effects of the legislation, including calls for scientific and conclusive proof of alleged effects is best left to the assessment of proportionality, infra. It is my view that legislation will be rationally connected to its intended objective if it (a) is designed to meet its objective, (b) is not arbitrary; and, (c) is based on assumptions which, logically applied, further the objective.

[103]   This view accepts and adopts the Supreme Court’s common-sense approach to defining the scope of the rational connection test. For example, in the recent case of M. v. H., Justices Cory and Iacobucci, speaking for the majority on the point, wrote simply that:

At the second stage of the s. 1 analysis, the focus shifts from the objective alone to the nexus between the objective of the provisions under attack and the means chosen by the government to implement this objective. As I have already stated, the means chosen include both the impugned provision and the omission in question. It falls to the party invoking s. 1 to demonstrate that there is a rational connection between the objective and the means.[106] [Citation omitted.]

[104]   In Thomson Newspapers, Gonthier J., dissenting on other grounds, described the common-sense view of the rational connection test in the following manner:

”The essence of rational connection is a causal relationship between the objective of the law and the measures enacted by the law. This is often a difficult matter to establish by evidence, and the Supreme Court of Canada has not always insisted on direct proof of the causal relationship.” … In RJR-MacDonald , supra, the Court unanimously agreed that a causal relationship between advertising and tobacco product consumption could be based either on common sense, reason, or logic (La Forest J., at para. 86, McLachlin J., at paras. 156-58, and Iacobucci J., at para. 184), even though the evidence may be admittedly inconclusive. In Butler, supra, Sopinka J. found, at p. 502, that it was “reasonable to presume” that there is a causal relationship between obscenity and harm to society. Similarly, in Ross, supra, La Forest J., writing for the Court, held, at para. 101, that it was “reasonable to anticipate” that there is a causal link between anti-Semitic activity by school teachers outside school and discriminatory attitudes within school.[107] [Citation omitted.]

[105]   In this case, the Trial Judge properly found that the legislation was rationally connected to its objectives despite a lack of empirical evidence. With regard to the first objective, the Trial Judge considered at length the expert evidence and held that there was a rational connection between disenfranchisement and the enhancement of civic responsibility and the rule of law. He seems to have accepted the evidence of Dr. Thomas Pangle and Dr. Christopher Manfredi, who testified that a legislative objective of enhancing civic responsibility and fostering respect for the rule of law was rationally connected to legislation which denounces disrespect for the process of law and for the social contract, and which restricts the franchise as a means of showing connection to the Canadian polity.[108]

[106]   I agree with the Trial Judge that the legislation is rationally connected to the government’s first objective. While it can be argued that all legislative action is meant to foster civic responsibility and enhance respect for the rule of law, a government may infringe a Charter right in order to foster objectives which are broadly stated. An abstract governmental objective is not, in and of itself, irrational. Undue breadth, of course, may give rise to difficulty at other stages of the test, but the rationality of an objective cannot be undermined by its breadth alone.

[107]   With respect to the second objective, the Trial Judge concluded at page 892 of his reasons that there was a rational connection between the disenfranchisement of prisoners and the objective of furthering the criminal sanction. He wrote:

While there is no empirical evidence to suggest that the disenfranchisement of prisoners reduces crime (either generally or specifically), or serves a morally educative function, or could operate as an effective punitive sanction, I find that a rational connection exists between the impugned provision and the stated objective of enhancing the criminal sanction. As an aid to punishment, the provision clearly imposes a sanction, and denounces bad conduct. In the present case, the sanction takes the form of a disenfranchisement, in addition to the loss of liberty. A fundamental democratic right has been removed for crimes committed, and its removal is clearly felt as a deprivation by Mr. Sauvé and Mr. Spence. It is also reasonable to conclude that a morally educative message is sent to offenders, and possibly to the general population, by the imposition of a sanction. Of course, the message may not be heard or understood, but that does not diminish the connection between the means and the second objective.

[108]   I concur with the Trial Judge and adopt his reasoning on this issue. In the result, I would find that the legislation is rationally connected to the pressing and substantial objectives which motivated it.

4. Does paragraph 51(e) of the CEA impair the Charter right in an appropriately minimal way?

[109]   The minimal impairment test seeks to determine whether the impugned state action impairs the right in an appropriately minimal way. The earliest cases demanded that Parliament use the “least restrictive means” to achieve its legislative objectives. Even in early cases, however, the Supreme Court sought to infuse flexibility into the Oakes test, realizing that a rigid application of a minimal impairment test would run the risk of improperly usurping the role of Parliament and “substitute judicial opinions for legislative ones as to the place at which to draw a precise line”.[109] Over time the Supreme Court also spoke of a “modified” section 1 test in situations where Parliament has chosen a balance between competing groups. Under the modified approach, the question is more appropriately whether Parliament could reasonably have chosen alternative means which would have impaired the right in question less or not at all but which would have achieved the identified objective as effectively. By contrast, it was said that where the state acts as a “singular antagonist” against an individual, courts must examine the impugned legislation more strictly.[110] In this case, the Trial Judge correctly considered several, more recent cases noting that Parliament need not adopt the absolutely least-intrusive means of attaining its objectives.[111]

[110]   In the recent case of Ross v. New Brunswick School District No. 15, La Forest J., writing for a unanimous court, summarized the test for minimal impairment in this fashion:

In RJR-MacDonald, supra, at p. 342, McLachlin J. reasoned that an impairment must be minimal to the extent that it impairs the right no more than is necessary. She stated:

The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement.[112]

[111]   In assessing minimal impairment, a court must therefore accord some deference to Parliament. In a case such as this, one matter which must be settled is the appropriate level of deference which this Court must show to Parliament on this issue.

[112]   In Thomson Newspapers, supra, Bastarache J. wrote (for a majority of the Court) that the context of each case would determine the amount of deference given by the courts to Parliament’s impairment of the Charter right.[113] He instructed that factors which determine the level of deference included the vulnerability of the group sought to be protected,[114] the presence of a logical inference refuting the existence of harm or social problem,[115] the opposition of group interests,[116] and whether the harm against which the legislation is intended to act is actual or potential.[117]

[113]   The context of each case must determine the level of deference to be shown to Parliament. While it is important to consider the state’s position as “singular antagonist”, that cannot and does not end the question of context.[118] For example, in RJR-MacDonald Inc., supra, the Supreme Court considered the constitutional validity of restrictions on tobacco advertising. Like this case, RJR-MacDonald Inc., supra, dealt with Parliamentary use of the criminal law power outside the Criminal Code. McLachlin J. reasoned that the “singular antagonist” test is not always easy to apply, and in many cases cannot in and of itself form a complete analysis of deference:

Related to context is the degree of deference which the courts should accord to Parliament. … For example, it has been suggested that greater deference to Parliament or the Legislature may be appropriate if the law is concerned with the competing rights between different sectors of society than if it is a contest between the individual and the state: Irwin Toy, supra, at pp. 993-94; Stoffman v. Vancouver General Hospital, …, at p. 521. However, such distinctions may not always be easy to apply. For example, the criminal law is generally seen as involving a contest between the state and the accused, but it also involves an allocation of priorities between the accused and the victim, actual or potential. The cases at bar provide a cogent example. We are concerned with a criminal law, which pits the state against the offender. But the social values reflected in this criminal law lead La Forest J. to conclude that “the Act is the very type of legislation to which this Court has generally accorded a high degree of deference” (para. 70). This said, I accept that the situation which the law is attempting to redress may affect the degree of deference which the court should accord to Parliament’s choice. The difficulty of devising legislative solutions to social problems which may be only incompletely understood may also affect the degree of deference that the courts accord to Parliament or the Legislature. As I wrote in Committee for the Commonwealth of Canada v. Canada, …, at p. 248, “some deference must be paid to the legislators and the difficulties inherent in the process of drafting rules of general application. A limit prescribed by law should not be struck out merely because the Court can conceive of an alternative which seems to it to be less restrictive”.[119] [Emphasis added; citations omitted.]

[114]   In the context of this case, the Parliament has chosen to deny the rights of a group of relatively powerless people—those serving prison sentences of two years or more—and must therefore be subject to a certain degree of scrutiny regarding its choice to do so. On the other hand, this statute represents an example of the state setting the ground rules for its electoral process. While the notion of ensuring a “decent” or “moral” electorate may have little place in today’s society, it is Parliament’s role to maintain and enhance the integrity of the electoral process.[120] Such considerations are by definition political and therefore warrant deference. This statute also represents an exercise of the criminal law power which is necessarily linked to the criminal sanction. It is my view that Parliament is entitled to a great deal of deference when it makes choices regarding penal policy. In R. v. Goltz, the Supreme Court held that a mandatory minimum sentence imposed by the British Columbia Motor Vehicle Act [R.S.B.C. 1979, c. 288] did not infringe the guarantee against cruel or unusual punishment found in section 12 of the Charter. Speaking for a majority of the Court, La Forest J. wrote that Parliament’s choices regarding the gravity of offences and the punishment which follows from conviction are entitled to considerable deference:

It is not for the court to pass on the wisdom of Parliament with respect to the gravity of various offences and the range of penalties which may be imposed upon those found guilty of committing the offences. Parliament has broad discretion in proscribing conduct as criminal and in determining proper punishment. While the final judgment as to whether a punishment exceeds constitutional limits set by the Charter is properly a judicial function, the court should be reluctant to interfere with the considered views of Parliament and then only in the clearest cases.[121]

[115]   It is also notable that the Supreme Court has upheld mandatory sentences as constitutionally valid,[122] and that upper limits on fixed term sentences were held inappropriate in certain cases. In R. v. M. (C.A.) [[1996] 1 S.C.R. 500], the Supreme Court held that a sentence beyond that fixed by the Criminal Code may, in the appropriate context, advance the traditional continuum of sentencing goals ranging from deterrence, denunciation and rehabilitation to the protection of society.[123]

[116]   I recognize that both Hugessen J.A. [in Belczowski] and Arbour J.A. [in Sauvé] felt that the predecessor to this statute, which disenfranchised all prisoners, should be strictly scrutinized. I agree completely that in the context of those cases strict scrutiny was warranted. Since those decisions, however, there has been considerable change, not only in the narrowing of the impugned legislation and in the application of the Oakes test, as discussed above, but also in the views of Canadians regarding the penal sanction. This Court can appropriately note that since 1992 Canada has seen two federal elections in which views of crime and punishment were important. Since 1992, Canada’s denunciation of crime and criminal behaviour has grown louder. The federal government has strengthened many aspects of the criminal law in an attempt to reflect the growing intolerance of crime in our communities. It is noteworthy that Parliament has also expended resources seeking alternatives to incarceration, and placing emphasis on victim’s rights. While it is important to remember that the Charter exists to protect vulnerable people from oppressive public moods, it is also important to be sensitive to legitimate changes in Parliamentary attitudes toward what is and is not sound penal policy. In the instant case, Parliament debated the measure in question vigorously and sought to respond to the constitutional concerns that this Court and the Ontario Court of Appeal raised with regard to the previous legislation. In its consideration of the former legislation, the Supreme Court had an opportunity to state that Parliament could not disenfranchise prisoners. It did not do so, choosing only to state that the former legislation was too broad. This left Parliament the option of enacting narrower legislation.

[117]   In this case, the Trial Judge recognized that some deference was warranted, but held that Parliament’s choice was seriously flawed by its lack of consideration of an individualized, court-based disenfranchisement. At page 897 of his reasons he wrote that:

The legislative history of paragraph 51(e) of the CEA displays virtually no consideration of a court based process where disqualification is considered as part of sentencing. What the legislative history does reveal, in somewhat vague terms, is an apparent desire to keep the matter out of the courts. In the House of Commons Debates, some reference was made to the issue of whether or not a criminal like Clifford Olsen should be permitted to vote. With disqualification on a case-by-case basis, a clearly indecent and immoral offender like Clifford Olsen could, as a consequence, be disenfranchised by sentence of the court.

[118]   The Trial Judge reasoned that courts are central to sentencing, and that judge-imposed disenfranchisement would be a less intrusive manner of achieving the legislative objective. The Trial Judge also reasoned that court-imposed disqualification would be more widely communicated than the current legislation:

As I shall discuss later, prisoner disenfranchisement is not well-known or visible in Canada. Certainly, any contemplated reform of the law, if pursued, could take this obvious consideration into account. The communication of sanctions to the public is the only obvious way for them to be effective. No doubt, minimizing disparity is an important goal, and perhaps only legal specialists can find their way through the maze of judge-made law; nevertheless, the public has a greater chance of being informed of the prisoner disenfranchisement through a court imposed disqualification, rather than under the current scheme.

The defendants’ experts have provided a number of thoughtful and compelling arguments supporting disenfranchisement. If Parliament decides to pass another law, there is no reason why these arguments could not inform the criteria that could be selected. A sentencing judge could take into account the nature of the crime and the personal circumstances of the accused in conjunction with the principles of sentencing: R. v. Goltz, supra. This process would, in my opinion, be a significantly less intrusive, and equally effective means of infringing a citizen’s democratic right to vote. If a judge is entrusted with the responsibility of taking away a person’s liberty, should he or she not also be charged with the responsibility of determining if disenfranchisement is warranted?[124]

[119]   With great respect for the thoughtful reasons of the Trial Judge, I believe he erred in coming to the conclusion that he did.

[120]   First, it is important to recall that this law was carefully tailored to affect only Canada’s most serious offenders. In order for a person to be disenfranchised, that person must be convicted beyond a reasonable doubt of a serious offence and be sentenced to prison for a period not less than two years. In our criminal justice system, a jail term of two years or more is a serious matter, normally leading to incarceration in a federal penitentiary.[125] On average, people who are sentenced to jail terms exceeding two years are guilty of serious offences like homicide, rape or robbery or are multiple offenders. For example, as of April 6, 1995, the average number of convictions per federal inmate was estimated at 29.5 convictions. A full discussion of these statistics follows, infra. Further, the disqualification only applies to those actually incarcerated after being sentenced to jail terms of two years or more. Those accused who are out on bail may vote in federal elections, as well as those who are out on parole, or those who have been convicted and are released pending appeal. While the earlier provision was struck down on the basis that the actual objectives were inappropriate and not rationally connected to the legislation, the same cannot be said of this provision. Parliament studied the matter and was aware of which crimes may yield sentences of two years or more (see the table reproduced below), and, listening to this Court and to the Ontario Court of Appeal, tailored a law which was aimed at disenfranchising only Canada’s most serious offenders.

[121]   Second, the Trial Judge himself noted that the legislative history reveals that Parliament, in tailoring this law, considered and rejected judge-imposed disenfranchisement, citing “an apparent desire to keep the matter out of the courts”.[126] It is obvious that the additional complexity which would be added to the sentencing exercise and the extra resources required to administer such a system was not attractive to Parliament. Further, the right to vote could depend on the effectiveness of counsel in their submissions on sentencing, something that is clearly unacceptable, considering the importance of section 3 in our society. That Parliament considered and rejected that option should be a signal to the courts that it was acting neither arbitrarily nor in haste. In my respectful view, Parliament need not examine the finest details of each and every option open to them in order to warrant deference. Nor must Parliament choose the absolutely least intrusive means of achieving a legislative goal, particularly where one objective of the law is to loudly denounce serious criminal behaviour. Before coming to this conclusion, Parliament expressly considered the Lortie Commission recommendation that prisoners serving a sentence of ten years or more be disenfranchised, following which it considered and rejected a motion to enact the disenfranchisement on sentences of five years or more, as well as a motion to repeal the disenfranchisement entirely. While it cannot be said that Parliament expressly considered disenfranchisement following conviction for a particular list of crimes, Parliament was aware that the provision they chose would attach to a finite number of crimes, i.e., those which can yield sentences of two years or more. All of these options were weighed and rejected. It cannot be said in this case that Parliament acted arbitrarily, or that Parliament insufficiently considered the matter, or that it chose an unreasonable solution to this difficult social problem.

[122]   Third, while the Harvey case, supra, is not squarely on point, its analysis of minimal impairment is instructive. In that case, the Supreme Court considered the constitutionality of a provision of the New Brunswick Elections Act which prohibited people from running in elections within five years of being convicted of electoral fraud. Writing for the majority, La Forest J. considered the “pre-determined” disqualification of persons and, as in this case, considered the argument that a fixed period of disqualification was unduly onerous because people would be affected based in part on chance, i.e., how many elections fell during their period of disqualification. He wrote that a fixed period of disqualification was appropriate:

The imposition of a five-year disqualification raises a more difficult issue. Accepting, as I do, that there is a rational connection between some term of disqualification and the desired objective, the question is reduced to what period of disqualification represents a minimal impairment of the appellant’s s. 3 Charter rights. It clearly makes sense that the appellant should not be allowed to run in the by-election that arises from him vacating his seat, but how much further should it go? In settling on a five-year disqualification the legislature has ensured that the appellant is ineligible to run in the next general election. While it is true that there could be several elections within that time, the legislature has chosen a fixed term, presumably because of its added certainty. In addition, a five-year disqualification provides for a time of cleansing, allowing the integrity of the electoral process to be renewed both in real terms and in the mind of the electorate.

This Court has on several occasions asserted its unwillingness to second-guess the legislature in choosing between acceptable options. In R. v. Edwards Books and Art Ltd., …, at p. 782, Dickson C.J. indicated that “[t]he courts are not called upon to substitute judicial opinions for legislative ones as to the place at which to draw a precise line”, and in RJR-MacDonald , supra, McLachlin J. had this to say, at p. 342:

The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement….

In the present instance I can see no reason why this Court should interfere with the balancing engaged in by the Legislature. A degree of deference is especially appropriate in this case where the impugned legislative provisions are aimed at transgressing members of the New Brunswick Legislative Assembly. Surely the members of that body are in the best position to choose between available options when it comes to deterring other members from breaching the trust that exists between them, the electorate, and the House as a whole.[127] [Emphasis added.]

[123]   Here too Parliament has sought to further electoral goals with a period during which the person convicted of the most serious crimes will be prohibited from participating in the law-making process. I can see no reason why this Court should declare invalid the balancing engaged in by Parliament in this case.

[124]   Fourth, the Trial Judge appears to have believed that this legislation can be characterized as an exercise in sentencing only, and he, consequently, held that sentencing is best carried out by sentencing judges. I cannot characterize this legislation exclusively as an attempt to add a Parliamentary sentence to that handed out by the judiciary. Parliament enacted this law in pursuit of objectives which are not so one-dimensional. There can be no debate that this provision, in addition to its electoral component, creates a consequence to criminal conviction.[128] That this provision creates a consequence to criminal conviction does not, however, transform it into a sentencing provision. The criminal sanction, generally speaking, is not limited only to sentencing. For example, the second edition of Clayton Ruby’s work, Sentencing, distinguishes orders which can be made by trial judges following conviction from criminal penalties which arise strictly by operation of law. Ruby writes:

These [judge-made] orders should be distinguished from penalties which arise strictly by operation of law. In such cases there is no jurisdiction in [sic] the trial judge to make an order at all. For example, in R. v. Berger ([1971] 1 O.R. 765 (C.A.)) the trial judge, pursuant to a conviction for having in possession goods unlawfully imported into Canada, purported to penalize the accused by an amount equal to the value of the goods, and to order forfeiture of the goods. The court said that the trial judge had no jurisdiction to make such orders; since these penalties occur by operation of law pursuant to the terms of the Customs Act they are not part of the sentence imposed by the Court.[129] [Emphasis added.]

[125]   Further, it is well settled that legislation may validly provide for a civil disability arising out of a criminal conviction. For example, in Provincial Secretary of Prince Edward Island v. Egan,[130] Rinfret J., in a provincial fishing licence case, wrote that:

The Provincial legislation in question in this case is, in pith and substance, within the classes of subjects assigned to the Provincial legislatures; it is licensing legislation confined to the territory of Prince Edward Island …. It is not an additional penalty imposed for a violation of the criminal law. It provides for a civil disability arising out of a conviction for a criminal offence …. It does not create an offence; it does not add to or vary the punishment already declared by the Criminal Code; it does not change or vary the procedure to be followed in the enforcement of any provision of the Criminal Code. It deals purely and simply with certain civil rights in the Province of Prince Edward Island.[131]

[126]   Civil consequences of a criminal act are not necessarily viewed as part of the criminal sentence. They neither create an offence nor add to or vary the punishment declared by the Criminal Code. Indeed it is settled that such consequences are not legally viewed as punishment.[132] That civil consequences arise from or are related to criminal activity does not diminish their legitimacy or efficacy.[133] While the difference between the civil consequence arising from conviction and the criminal sanction may seem obscure at times, Parliament and provincial legislatures may enact legislation giving rise to such a consequence, even where such a consequence resembles a criminal sanction.[134]

[127]   Fifth, the Trial Judge’s reasoning, effectively requiring individualized disenfranchisement, wrongly neglects the individualized elements of this sanction. There is proportionality built into this legislation as it impacts on the particular offender. In general, elections are held every four or five years. If offenders are incarcerated for only two years, they may not even actually be deprived of their vote, depending, of course, on the timing of any election. On the other hand, a person incarcerated for twenty years is likely to miss voting in several elections. While the connection between the length of the sentence and the severity of the deprivation is not perfect, there is a rational range here: those who are convicted of the most serious crimes—and handed the most serious sentences—will be disqualified from voting for longer than those who are sentenced only for two years.

[128]   It is also important not to overlook that this legislation does not apply to persons on parole. Parliament has made a choice that it is only while in prison that a person is disenfranchised. Once a person is released on parole, that person may again vote in federal elections. Thus the right to participate in electing our representatives is restored to those convicted prisoners who are considered ready to return to life outside prison. To the extent that the deprivation of the right to vote is a meaningful one, the promise of automatic return of that right to persons on parole serves as an additional incentive for convicted prisoners to behave well while imprisoned, and to be rehabilitated.

[129]   In conclusion, this prohibition is a hybrid which possesses elements of the criminal sanction as well as elements of civil disability based on electoral law. While it is linked to the exercise of the criminal law power, the provision also pursues valid electoral goals. With respect, the Trial Judge impoverished the provision when he reasoned that it was merely a supplementary sentencing provision. Parliament, basing itself on electoral policy, is entitled to add civil consequences to the criminal sanction in subtle, multi-dimensional ways. This legislation is just that—a complex mixture of criminal and electoral law which creates a disqualification following criminal conviction, a type of measure that is not unknown to federal legislation.[135] Parliament may pass the scrutiny of section 1 of the Charter by considering alternative means and choosing a law which “falls within a range of reasonable alternatives”.[136] As McLachlin J. explained in terms most helpful here:

… some deference must be paid to the legislators and the difficulties inherent in the process of drafting rules of general application. A limit prescribed by law should not be struck out merely because the Court can conceive of an alternative which seems to it to be less restrictive.[137]

5. Is paragraph 51(e) of the CEA proportional?

[130]   The proportionality aspect of the Oakes test has undergone something of a reformulation in recent times. This reformulation, instructing courts to weigh the salutary and deleterious effects of the legislation, does not alter the overriding instruction that the Oakes test must be a flexible one, based primarily on the context of the case at bar. In the Dagenais case, Lamer C.J. summarized his reformulation of the proportionality analysis as follows:

I would, therefore, rephrase the third part of the Oakes test as follows: there must be a proportionality between the deleterious effects of the measures which are responsible for limiting the rights or freedoms in question, and there must be proportionality between the deleterious and salutary effects of the measures.[138] [Emphasis in original.]

[131]   In the recent case of Thomson Newspapers, Bastarache J. expanded on the analysis between the deleterious and salutary effects in the context of the Oakes test generally. He wrote that:

The focus of the first and second steps of the proportionality analysis is not the relationship between the measures and the Charter right in question, but rather the relationship between the ends of the legislation and the means employed. Although the minimal impairment stage of the proportionality test necessarily takes into account the extent to which the Charter right is infringed, the ultimate standard is whether the Charter right is impaired as little as possible given the validity of the legislative purpose. The third stage of the proportionality analysis provides an opportunity to assess, in light of the practical and contextual details which are elucidated in the first and second stages, whether the benefits which accrue from the limitation are proportional to its deleterious effects as measured by the values underlying the Charter.[139]

[132]   Thus the third part of the proportionality test involves an inquiry into whether the deleterious effects of the measures are proportional to the significance of the objectives and the beneficial effects of the measure. This inquiry is always a difficult and imprecise process. The context of each case is always paramount, and there can be no rigid test for an assessment of proportionality.[140] A court weighing the proportionality between the measures and the pressing and substantial objective of the provision must therefore take into account factors such as the significance and nature of the state’s objectives; the availability of evidence; evidence of the effectiveness of the objectives; the benefits of the legislation; the nature of the Charter right and the extent to which the Charter right is interfered with; and, the manner, e.g., direct or indirect, in which the Charter right is compromised.[141]

[133]   In the Harvey case, supra, the Supreme Court, despite the above jurisprudence, considered primarily whether the means chosen by the legislation were proportional to the objectives in question, and did not specifically weigh the salutary and deleterious effects.[142] While this is by no means determinative of the appropriate analysis of proportionality in this case, it further highlights that the context of the particular case is paramount in the Oakes analysis.

[134]   In the context of this case, this statute represents, in large part, an exercise of the criminal law power which is necessarily linked to the criminal sanction. In passing this legislation, however, the state has disenfranchised a group of citizens for the period that they are incarcerated in federal prison. Is this proportional? As a preliminary matter, it is hard to speak of salutary effects in the context of the penal sanction, especially in an age where there is little evidence proving that the penal sanction is effective in reducing or deterring crime, or in reducing recidivism. Put another way, it is very difficult for the government to demonstrate empirically salutary effects in this case. This cannot, however, be controlling. If it were, all penal sanctions, including jail sentences, would be called into question, for it is by no means easy to demonstrate their efficacy in reducing crime scientifically.

[135]   There can be no doubt that, in addition to electoral considerations, the main motivations in passing this law were the retributive and denunciatory aspects of the penal sanction. While many penologists may disapprove of these goals, these are important and legitimate objectives for Parliament, in its wisdom, to pursue. The courts cannot prevent Parliament from proportionately compromising Charter rights in the name of denouncing crime, even if they disagree with Parliament’s penal philosophy.

[136]   Despite the difficulty in discussing the salutary effects of retribution and denunciation, there are salutary effects of this legislation as well as valid objectives which were identified to this Court.

[137]   The main salutary effect of this legislation is a complex but important one. The legislation dramatically expresses the sense of societal values of the community in relation to serious criminal behaviour and the right to vote in our society. It is not merely symbolic. This legislation sends a message signalling Canadian values, to the effect that those people who are found guilty of the most serious crimes will, while separated from society, lose access to one of the levers of electoral power. This is an extremely important message, one which is not sent by incarceration alone. Incarceration is essentially separation from the community. Incarceration alone signals a denunciation of the offender’s anti-societal behaviour and indicates society’s hope for rehabilitation through separation from the community. Incarceration by itself, however, leaves those convicted of serious crimes free to exercise all the levers of electoral power open to all law-abiding citizens. This maintains a political parity between those convicted of society’s worst crimes and their victims. Disqualification from voting, however, signals a denunciation of the criminal’s anti-societal behaviour and sends the message that those people convicted of causing the worst forms of indignity to others will be deprived of one aspect of the political equality of citizens—the right to vote.[143] It can be said that, in this context, “kindness toward the criminal can be an act of cruelty toward his victims, and the larger community”.[144]

[138]   In this way, the enactment of the measure is itself a salutary effect, and the disqualification serves a messaging function. That penal legislation can validly signal our country’s values was confirmed by the Supreme Court in R. v. Butler, where Sopinka J. wrote for the Court that:

On the other hand, I cannot agree with the suggestion of the appellant that Parliament does not have the right to legislate on the basis of some fundamental conception of morality for the purposes of safeguarding the values which are integral to a free and democratic society. As Dyzenhaus … writes:

Moral disapprobation is recognized as an appropriate response when it has its basis in Charter values.

As the respondent and many of the interveners have pointed out, much of the criminal law is based on moral conceptions of right and wrong and the mere fact that a law is grounded in morality does not automatically render it illegitimate.[145]

[139]   This legislation proclaims that values of civic responsibility are important to Canadians. The signal itself is an important benefit of the law.[146] The signal itself is a double signal, a message about the community’s view of crime and a repudiation of the indignity perpetrated on victims of crime. Where someone, by committing a serious crime, evinces contempt for our basic societal values, their right to vote may be properly suspended. Indeed, not to do so undermines our democratic values.[147]

[140]   This view was presented at trial by the Crown’s expert witnesses. For example, in his expert’s report, Dr. Thomas Pangle explains the provision as a public enunciation of the community standards of civic responsibility:

For honors and penalties, advantages and disadvantages, qualifications and disqualifications set in the legal code are to be seen not only as rewards and punishments—whether aimed at retribution, or deterrence, or rehabilitation—but also, in some cases (as in the electoral qualifications) more importantly, as the community’s mutual civic education, through the solemn public enunciation of the community standards of civic responsibility. Our laws tell us our story.[148]

[141]   In context, this legislation puts forward a statement of principles by which Canadians live. That is a valid role which Parliament may play. The formal enactment of these principles itself is as important as any tangible effects that the law may have. This is not an exercise in choosing a “virtuous” or “decent” electorate. As Dr. Pangle indicated in his testimony, a statement of values exists as a message to prisoners that their serious criminal conduct has, as its consequence, withdrawal of one of the levers of political power. Dr. Pangle further indicated in his testimony that a temporary disenfranchisement reminds the prisoner of the standards which we as a society demand before all of the levers of political power will be returned.[149] Dr. Manfredi, testifying for the Crown, offered a similar view of the disenfranchisement, noting a moral education function created by the link between good citizenship—which includes, inter alia , not committing the most heinous crimes imaginable—and the deprivation of one of the levers of political power. Dr. Seymour Martin Lipset, pointing out Canada’s emphasis on community, suggested in his testimony that the provision is congruent with Canada’s political culture.[150]

[142]   Thus the suspension of the right to vote—and not necessarily the actual loss of a vote—becomes significant. If a citizen commits one or more of the most serious crimes, crimes which repudiate Canada’s sense of community and which demolish the dignity of their victims, then that citizen’s behaviour will be denounced through the removal of the vote, one of the levers of political power.[151] The legislation becomes a medium, and the medium sends a message.

[143]   Second, there was some evidence of a particular salutary effect presented to this Court, that is, that disenfranchisement is a meaningful sanction which is noticed by offenders. In his testimony before the Trial Division, the plaintiff Aaron Spence was asked by Crown counsel about the effect on him of the disenfranchisement. Mr. Spence was aware that many rights are withdrawn from prisoners, and indicated that the loss of the right to vote was meaningful:

Q. I understand. But to answer my question, sir, is you have reflected on the consequence of your sentence, and this sentence is pretty unpleasant.

A. Without a doubt.

Q. Yes. And, in fact, one of the reasons, one of the reasons it is unpleasant, Mr. Spence, is because your time here is spent without many of the things you enjoy on the outside, correct? Isn’t that kind of an obvious question?

A. Well, not really, it is not an obvious question, because within this institution, I have come to the terms to such an extent within my holistic healing that I am happy with what I have.

Q. Mr. Spence, let me ask the question of you again. Here in prison, during the sentence which you have already admitted is quite uncomfortable and unpleasant, you are, in fact, deprived of many things that you would otherwise have on the outside, it is a very simple question; isn’t that correct?

A. Correct.

Q. Yes?

A. Yes.

Q. And some of those things are, in fact, valuable things like, for instance, the right to vote, yes?

A. Well, yeah. I would say there is a lot of things that is affected, and like I said earlier as an answer to one of your earlier questions, one of the most important things to me as a person, not only as an aboriginal person, but the things that are affecting my family on the outside, that is what hurts a lot too. My family, my friends, my family circle.

Q. You, yourself, as a sentenced person in Stony Mountain Institution, Stony Mountain Penitentiary, you are deprived of certain things you would like on the outside, simple question, yes?

A. Other than, other than having my family here on a daily basis, I would have to disagree with you to a certain extent. This ain’t Club Fed min [sic] you, but I am saying within my holistic being, my journey and being inside here, I have come to live and accept with myself. I love myself. I sleep good every night. This place is a living hell, but I sleep good every night.

Q. The fact that you are deprived of the right to vote doesn’t bother you?

A. It does, it does bother me.

Q. The answer to my question, then, Mr. Spence, unless I misunderstood your answer, is that you are, in fact, deprived of certain things that you think are valuable?

A. Definitely.

Q. Like the right to vote?

A. That is true.[152]

[144]   Lastly, this legislation can be seen as a gentler, more humane alternative to the alternative of additional incarceration. Parliament could easily have chosen, instead of denying the vote to serious offenders, to further denounce crime by increasing the length of all criminal sentences of two years or more. Indeed it was the position of Mr. O’Connor, counsel for the respondent Sauvé, that had Parliament wanted to further punish offenders, it should have increased the length of sentences. I disagree. This Court can appropriately take judicial notice of the fact that Parliament must be allowed to search for better alternatives to further incarceration. In an age where the efficacy of lengthy imprisonment is being questioned, our legislators can and must look for other ways of denouncing crime. In the battle against crime, Courts cannot limit Parliament to a single punitive tool. Rather than increasing all sentences of two years or more, Parliament has chosen an alternative to further incarceration by denying the vote to Canada’s most serious criminals during the period in which they serve their sentences.

[145]   Contrary to these salutary effects, the sole deleterious effect of this legislation is the withdrawal of the Charter-guaranteed right to vote. While this deprivation is serious—and to the extent that one is disqualified during a particular election, total—several facts were brought to the attention of this Court which mitigates its deleterious nature. The correlation between the length of the sentence and the length of the disqualification, as discussed above, is individualized to an extent. Further, a report prepared for the Crown by expert witness Dr. Colin Meredith shows that 75% of prisoners incarcerated in federal penitentiaries are serving sentences of 5 years or less. Under normal circumstances, these people will miss voting at only one election.[153] Dr. Meredith’s report produced statistics showing that this legislation—targeted to disqualify the most serious offenders from voting—is meeting its goal. Dr. Meredith produced a table of the number and percentage of inmates in federal penitentiaries serving time for particular offences. This table is reproduced here:

Offence		Number	Percent
Murder		2,041	14.4
Attempted murder		451	3.2
Manslaughter		780	5.5
Sexual assault		1,236	8.7
Sexual offences involving children		574	4.0
Other sexual offences		961	6.8
Assaults		2,914	20.6
Kidnap		1,129	8.0
Robbery		4,633	32.7
Firearms offences		1,488	10.5
Prison breach		221	1.6
Arson		129	0.9
Break and enter		1,733	12.2
Fraud		345	2.4
Trafficking		1,606	11.3
Importing/exporting illegal narcotics		321	2.3
Cultivation of illegal narcotics		37	0.3
Possession of property obtained by certain offences		61	0.4
Total inmates		14,179
	Percentages do not sum to 100 because they are based on counts of offences rather than counts of inmates.

[146]   Further, Dr. Meredith concluded that the total number of convictions for the sample of individual federal inmates was found to be, on average, 29.5 convictions.[154] This suggests that the provision accurately targets multiple offenders. Dr. Meredith also found that the average number of convictions registered prior to the most recent conviction for the sample of federal inmates was found to be 25.2 convictions.[155] This suggests that the provision accurately targets not only serious criminals but repeat offenders.

[147]   The Trial Judge in this case considered that, while the salutary effects of the provision were not ascertainable, the deleterious effects were obvious. He wrote that the public’s ignorance of the disqualification annulled any salutary effects which the provision might have.[156]

[148]   With great respect for the Trial Judge, I believe that he erred in focussing on the public awareness of the provision as a proxy for the efficacy of its salutary effects. As I have already pointed out, one cannot speak of salutary effects in the context of this legislation as one might in the pursuit of more conventional objectives. Further, there are salutary effects, discussed above, which are not linked to public awareness of the provision—the provision loudly denounces crime, without increasing the length of jail terms.

[149]   It was submitted to this Court that public ignorance of this provision nullifies its proportionality. On reflection, I believe that argument misses the mark. While I am willing to assume that this disqualification is not widely known amongst the public, it cannot be assumed that sentencing judges and defence counsel do not know about this disqualification. In fact, it is reasonable to expect that judges know all the consequences of the sentences they hand out. Defence counsel are able to mention these in their sentencing submissions. Unlike the previous provision, which disqualified all prisoners and was therefore punishing imprisonment generally, under this provision offenders are disqualified from voting only after they have been indicted, convicted beyond a reasonable doubt of a serious crime and have received a sentence of two years or more. It is also the case that, in order to be disenfranchised, the prisoner must have exhausted all appeals, including sentencing appeals, and must have begun serving a prison term. Sentencing judges can be presumed to consider and weigh this consequence before handing down sentences of two years or more.

[150]   As a final note, it is important to remember that the Charter grants convicted persons two protections which directly challenge Parliament’s unfettered right to fix the criminal sanction, namely that no one may be deprived of liberty except in accordance with the principles of fundamental justice and that no one shall be subjected to cruel or unusual punishment. Paragraph 51(e) was not challenged under either of these provisions. I have no doubt that, if either of these sections were thought to be applicable, they would have been argued by the very capable counsel in this case.

[151]   Thus, viewed as a civil consequence imposed as an alternative to additional incarceration which attaches to the most serious sentences for the most serious crimes, it must be concluded that this measure is proportional. To conclude otherwise would be to challenge Parliament’s right to utilize the law to enhance civic responsibility and to establish consequences which express Canada’s abhorrence of serious crime. I would, therefore, hold that paragraph 51(e) of the CEA is a reasonable limit demonstrably justified in a free and democratic society within the meaning of section 1 of the Charter.

VI.       Does paragraph 51(e) of the CEA violate section 15 of the Charter?

[152]   While in the past there has been some variation in the views of the Supreme Court justices regarding the proper approach to subsection 15(1) of the Charter, these differences of opinion have diminished over time,[157] and finally have been laid to rest in the recent decision of Law v. Canada (Minister of Employment and Immigration),[158] a turning point in equality jurisprudence in Canada. In Law, Iacobucci J. synthesized the divergent views regarding the proper application of subsection 15(1) of the Charter into a new approach, unanimously accepted by his colleagues. Law is now the starting point for any analysis of discrimination under the Charter.

[153]   The proper approach to subsection 15(1) was summarized by Iacobucci J. at paragraph 39 [pages 523-524] of his reasons:

In my view, the proper approach to analyzing a claim of discrimination under s. 15(1) of the Charter involves a synthesis of these various articulations. Following upon the analysis in Andrews, supra, and the two-step framework set out in Egan, supra, and Miron, supra, among other cases, a court that is called upon to determine a discrimination claim under s. 15(1) should make the following three broad inquiries. First, does the impugned law (a) draw a formal distinction between the claimant and others on the basis of one or more personal characteristics, or (b) fail to take into account the claimant’s already disadvantaged position within Canadian society resulting in substantively differential treatment between the claimant and others on the basis of one or more personal characteristics? If so, there is differential treatment for the purpose of s. 15(1). Second, was the claimant subject to differential treatment on the basis of one or more of the enumerated and analogous grounds? And third, does the differential treatment discriminate in a substantive sense, bringing into play the purpose of s. 15(1) of the Charter in remedying such ills as prejudice, stereotyping, and historical disadvantage? The second and third inquiries are concerned with whether the differential treatment constitutes discrimination in the substantive sense intended by s. 15(1).

[154]   There are, accordingly, not unlike the position before Law, three questions to be answered by a party seeking to prove a violation under subsection 15(1) of the Charter. First, the claimant must show that the impugned legislation or state action—or its effects—draws a distinction between the claimant and others. Second, the claimant must show that the distinction has been drawn on an enumerated or analogous ground.[159] Third, the claimant must show that the distinction drawn on the basis of an enumerated or analogous ground is discriminatory in its nature.

[155]   The reformulation of the section 15 analysis that occurred in Law v. Canada (Minister of Employment and Immigration) did not affect the criterion that, in order to violate the Charter, any impugned distinction must be made on the basis of an enumerated or analogous ground. In M. v. H., Justices Cory and Iacobucci, speaking for the Court, wrote that:

Not every legislative distinction is discriminatory. Before it can be found that it gives rise to discrimination, it must be shown that an equality right was denied on the basis of an enumerated or analogous ground, and that this differential treatment discriminates “in a substantive sense, bringing into play the purpose of s. 15(1) of the Charter“: Law, supra, at para. 39 (emphasis in original).[160]

[156]   The respondents argue that the Trial Judge erred when he dismissed their claims under section 15 of the Charter. I am not in complete agreement with the reasoning of the Trial Judge on this matter. Specifically, I have concerns about the Trial Judge’s reasoning that since all disenfranchised people suffer to the same degree, then there can be no discrimination.[161] In my view, state action which is facially neutral may constitute discriminatory behaviour in appropriate contexts. Nonetheless, I am not persuaded that the Trial Judge came to the wrong conclusion on this issue.

[157]   In my view, the status of prisoner does not constitute an analogous ground for purposes of section 15 of the Charter in this case, and I would therefore conclude that there has been no violation of section 15 of the Charter here. Several cases have already addressed this matter. In the case of Jackson v. Joyceville Penitentiary, MacKay J. came to the same conclusion, writing that:

In this case section 41.1 of the Regulations does provide for treatment of prison inmates on a basis different from that of most, if not all, other individuals in Canada and thus may be said to derogate from their right to equality before the law. Yet I am not persuaded that this difference or derogation is discriminatory in the sense provided for in subsection 15(1). It does not seem to me to be related to any of the enumerated and prohibited grounds, or analogous grounds, which concern personal characteristics. The difference in treatment here for prison inmates as a group, arises not from personal characteristics but from past courses of conduct amounting to criminal activities against society.[162] [Emphasis added.]

[158]   In his judgment in Belczowski v. Canada, Strayer J. (as he then was) wrote that, in his view, prisoners did not constitute an analogous group warranting protection under the Charter’s equality provision:

I believe it is now clear since decisions such as R. v. Turpin … and Reference Re Workers’ Compensation Act … that for there to be “discrimination” as prohibited by subsection 15(1) the grounds of discrimination must be those specified in subsection 15(1) or others analogous thereto. I am unable to conclude that a law applied to the plaintiff to his disadvantage by reason of the circumstance that he has committed a crime and is imprisoned under lawful sentence amounts to discrimination on some ground analogous to those specified in subsection 15(1).

I therefore conclude that paragraph 51(e) of the Canada Elections Act like its predecessor paragraph 14(4)(e) is not inconsistent with section 15 of the Charter.[163]

[159]   The Tax Court of Canada has considered the issue of whether prison inmates constitute an analogous group for purposes of subsection 15(1) of the Charter on three separate occasions. On each occasion the Tax Court has answered the question in the negative. In 1991, in McKinnon (R.J.) v. M.N.R., Sobier J.T.C. quoted with approval the reasoning of Strayer J., supra.[164] In 1995, Mogan J.T.C. of the Tax Court of Canada considered the matter in Armstrong v. R. He also concurred with the reasoning of Strayer J. in Belczowski, supra, writing that the individual prisoner is neither part of a “discrete and insular minority”, nor is he imprisoned because of some “subjective distinguishing mark”.[165] Finally, in 1996, in Mulligan v. Canada, Bowie J.T.C. concurred with the reasoning of Mogan J.T.C. in Armstrong v. R., following which he noted that prisoners “do not by any stretch of the imagination” meet the standards set out by the Supreme Court for inclusion in an analogous group.[166]

[160]   In Olson v. Canada, Heald D.J., sitting as a designated judge of the Trial Division, again held that prisoners were not entitled to claim equal protection under subsection 15(1) of the Charter:

However, the third step of the analysis presents a problem for the plaintiff. It requires that the distinction be based on an irrelevant personal characteristic that is either enumerated in subsection 15(1) or is analogous thereto. The jurisprudence has established that differential treatment of prison inmates is not because of a personal characteristic, rather it arises from “past courses of conduct amounting to criminal activities against society”. Accordingly, since the third step of the Egan analysis has not been met, it follows that the plaintiff’s claim of infringement under subsection 15(1) of the Charter must fail.[167] [Footnote omitted.]

[161]   Finally, in the recent case of Alcorn v. Canada (Commissioner of Corrections), Richard A.C.J. agreed that prisoners were not a group analogous to those enumerated in subsection 15(1) of the Charter. He accepted the reasoning of the Trial Judge in this case and wrote that:

Regarding the second requirement “whether this denial constitutes discrimination on the basis of an enumerated or analogous ground”, I am not satisfied that the discrimination invoked by the applicants is “based on grounds relating to personal characteristics of the individual or group”.

…

In any event, I reiterate the position of this Court that prisoners and/or inmates do not constitute an analogous group under section 15 of the Charter as stated in Sauvé v. Canada.[168]

[162]   Past jurisprudence is unanimous that prisoners do not constitute a group analogous to those enumerated under subsection 15(1) of the Charter.

[163]   All of these decisions are influential, but not technically binding on this Court, and, as recent Supreme Court jurisprudence has given new instructions regarding the test for identifying an analogous ground, I shall briefly summarize the developments on this issue, confirming my view that prisoners do not constitute an analogous group for purposes of subsection 15(1) of the Charter.

[164]   The Supreme Court considered the enumerated and analogous grounds in the recent case of Corbiere v. Canada (Minister of Indian and Northern Affairs), the reasons for which were filed on the same day as M. v. H., supra. In that case, McLachlin and Bastarache JJ. summarized the purpose underlying the analogous grounds in this way:

The enumerated grounds function as legislative markers of suspect grounds associated with stereotypical, discriminatory decision making. They are a legal expression of a general characteristic, not a contextual, fact-based conclusion about whether discrimination exists in a particular case. As such, the enumerated grounds must be distinguished from a finding that discrimination exists in a particular case. Since the enumerated grounds are only indicators of suspect grounds of distinction, it follows that decisions on these grounds are not always discriminatory; if this were otherwise, it would be unnecessary to proceed to the separate examination of discrimination at the third stage of our analysis discussed in Law, supra, per Iacobucci J.

The same applies to the grounds recognized by the courts as “analogous” to the grounds enumerated in s. 15. To say that a ground of distinction is an analogous ground is merely to identify a type of decision making that is suspect because it often leads to discrimination and denial of substantive equality. Like distinctions made on enumerated grounds, distinctions made on analogous grounds may well not be discriminatory. But this does not mean that they are not analogous grounds or that they are analogous grounds only in some circumstances. Just as we do not speak of enumerated grounds existing in one circumstance and not another, we should not speak of analogous grounds existing in one circumstance and not another. The enumerated and analogous grounds stand as constant markers of suspect decision making or potential discrimination. What varies is whether they amount to discrimination in the particular circumstances of the case.[169] [Emphasis added.]

[165]   Having identified that the analogous grounds exist in order to identify suspect decision-making, the majority opinion continued to explain the method of identification of analogous grounds. McLachlin and Bastarache JJ. wrote:

What then are the criteria by which we identify a ground of distinction as analogous? The obvious answer is that we look for grounds of distinction that are analogous or like the grounds enumerated in s. 15—race, national or ethnic origin, colour, religion, sex, age, or mental or physical disability. It seems to us that what these grounds have in common is the fact that they often serve as the basis for stereotypical decisions made not on the basis of merit but on the basis of a personal characteristic that is immutable or changeable only at unacceptable cost to personal identity. This suggests that the thrust of identification of analogous grounds at the second stage of the Law analysis is to reveal grounds based on characteristics that we cannot change or that the government has no legitimate interest in expecting us to change to receive equal treatment under the law. To put it another way, s. 15 targets the denial of equal treatment on grounds that are actually immutable, like race, or constructively immutable, like religion. Other factors identified in the cases as associated with the enumerated and analogous grounds, like the fact that the decision adversely impacts on a discrete and insular minority or a group that has been historically discriminated against, may be seen to flow from the central concept of immutable or constructively immutable personal characteristics, which too often have served as illegitimate and demeaning proxies for merit-based decision-making.[170] [Emphasis added.]

[166]   Applying the law to this case, I cannot describe one’s status as a prisoner as “a personal characteristic that is immutable or changeable only at unacceptable cost to personal identity”. Imprisonment is neither immutable nor unchangeable; for all but a few prisoners it is a status that is meant to change over time. Further, it cannot be said that “the government has no legitimate interest in expecting” prisoners to change in order “to receive equal treatment under the law”. In fact, the contrary is true—the government has every reason to expect convicted criminals to change their behaviour in order to achieve equal treatment under the law. That is the very reason for imprisonment. I cannot conclude, therefore, that prisoners constitute an analogous ground under subsection 15(1) of the Charter.

[167]   I am bolstered in this conclusion by the existence of section 12 of the Charter, which prohibits the state from inflicting cruel or unusual punishment. That section reads as follows:

12. Everyone has the right not to be subjected to any cruel and unusual treatment or punishment.

[168]   The existence of section 12 of the Charter indicates that prisoners, like everyone else, are to be treated humanely, and are not to be subject to disproportionate punishment. The Charter, therefore, has a provision expressly targeted to protect prisoners’ condition. That provision recognizes that, in a way, prisoners are subject to unequal treatment and significant deprivation, but it seeks to ensure that such deprivation is proportionate. While there are other contexts in which a person who is a prisoner might take advantage of subsection 15(1) of the Charter, it is my conclusion that prisoners per se do not constitute a group analogous to those enumerated in subsection 15(1) of the Charter for purposes of this situation.

[169]   Finally, an alternative argument was made before this Court that imprisonment should be recognized as an analogous ground because Aboriginal peoples make up a disproportionate percentage of prisoners. I am not persuaded by this argument. First, since, according to the data offered, 1,837 Aboriginal people are disenfranchised by this law,[171] it cannot be said that the over-representation of Aboriginal peoples in the prison system adversely affects the political expression of Aboriginal peoples generally, as there are over six hundred thousand registered Aboriginal people in Canada.[172] If Aboriginal people generally, or a particular group of Aboriginal people, could show that disenfranchisement effectively and adversely compromised their political expression, a constitutional exemption from the operation of paragraph 51(e) of the CEA might conceivably be justified. This has not been done. Second, it cannot be said that the over-representation of Aboriginal peoples in prison is so overwhelming as to justify a conclusion that a law aimed at prisoners is de facto a law aimed at Aboriginal peoples. If the over-representation of Aboriginal peoples in prisons reaches a level where it could be said that a law aimed at prisoners was, de facto, a law aimed at Aboriginal peoples, then constitutional exemption from the operation of paragraph 51(e) of the CEA might be considered. That is not the case here. Consequently, I cannot accept the arguments made by the respondents in this regard.

VI.       Disposition

[170]   For all the foregoing reasons, I would conclude that while paragraph 51(e) of the CEA violates section 3 of the Charter, it is a reasonable limit demonstrably justified in a free and democratic society within the meaning of section l of the Charter. Furthermore, paragraph 51(e) of the CEA does not violate paragraph 15(l) Charter. In the result I would allow the appeal and dismiss the cross-appeal. Because of the history and importance of the constitutional issue involved in this appeal and because of the desirability of having it resolved there should be no order as to costs.

---