Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 SCR 519

Sauvé v. Canada (Chief Electoral Officer), [2002] 3 S.C.R. 519, 2002 SCC 68

Richard Sauvé                                                                                                  Appellant

v.

The Attorney General of Canada, the Chief Electoral Officer

of Canada and the Solicitor General of Canada                                         Respondents

and between

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                                                                                 Appellants

v.

The Attorney General of Canada                                                                 Respondent

and

The Attorney General for Alberta, the Attorney General of

Manitoba, the Canadian Association of Elizabeth Fry Societies,

the John Howard Society of Canada, the British Columbia Civil

Liberties Association, the Aboriginal Legal Services of Toronto Inc.

and the Canadian Bar Association                                                               Interveners

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

Neutral citation:  2002 SCC 68.

File No.:  27677.

2001:  December 10; 2002: October 31.

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

on appeal from the federal court of appeal

Constitutional law — Charter of Rights  — Right to vote — Prisoners —  Canada Elections Act provision disqualifying persons imprisoned in correctional institution serving sentences of two years or more from voting in federal elections —  Crown conceding that provision infringes right to vote — Whether infringement justified — Canadian Charter of Rights and Freedoms, ss. 1 , 3  — Canada Elections Act, R.S.C. 1985, c. E‑2, s. 51(e).

Constitutional law — Charter of Rights  — Equality rights — Prisoners —  Canada Elections Act provision disqualifying persons imprisoned in correctional institution serving sentences of two years or more from voting in federal  elections —  Whether provision infringes equality rights — Canadian Charter of Rights and Freedoms, s. 15(1)  — Canada Elections Act, R.S.C. 1985, c. E-2, s. 51(e).

Elections — Disqualifications of  electors — Prisoners — Canada Elections Act provision disqualifying persons imprisoned in correctional institution serving sentences of two years or more from voting in federal  elections — Whether provision constitutional — Canadian Charter of Rights and Freedoms, ss. 1 , 3 , 15(1)  — Canada Elections Act, R.S.C. 1985, c. E-2, s. 51(e).

Section 51 (e) of the Canada Elections Act denies the right to vote to “[e]very person who is imprisoned in a correctional institution serving a sentence of two years or more.”  The constitutionality of s. 51 (e) was challenged on the grounds that it contravenes ss. 3  and 15(1)  of the Canadian Charter of Rights and Freedoms  and is not demonstrably justified under s. 1.  In the Federal Court, Trial Division, the Crown conceded that s. 51( e )  infringes the right to vote guaranteed by s. 3  of the Charter  and the trial judge found that the infringement was not justified under s. 1  of the Charter .  The majority of the Federal Court of Appeal set aside the decision and upheld the constitutionality of s. 51( e ) .  The court held that the infringement of s. 3 was justifiable in a free and democratic society and that s. 51( e )  did not infringe the equality rights guaranteed by s. 15(1)  of the Charter .

Held (L’Heureux‑Dubé, Gonthier, Major and Bastarache JJ. dissenting): The appeal should be allowed.

Per McLachlin C.J. and Iacobucci, Binnie, Arbour and LeBel JJ.:  To justify the infringement of a Charter  right under s. 1, the government must show that the infringement achieves a constitutionally valid purpose or objective, and that the chosen means are reasonable and demonstrably justified.  The government’s argument that denying the right to vote to penitentiary inmates requires deference because it is a matter of social and political philosophy is rejected.  While deference may be appropriate on a decision involving competing social and political policies, it is not appropriate on a decision to limit fundamental rights.  The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside.  Limits on it require not deference, but careful examination.  The framers of the Charter  signaled the special importance of this right not only by its broad, untrammeled language, but by exempting it from legislative override under s. 33 's notwithstanding clause.  The argument that the philosophically‑based or symbolic nature of the objectives  in itself commands deference is also rejected.  Parliament cannot use lofty objectives to shield legislation from Charter  scrutiny. Here, s. 51( e )  is not justified under s. 1  of the Charter .

The government has failed to identify particular problems that require denying the right to vote, making it hard to conclude that the denial is directed at a pressing and substantial purpose.  In the absence of a specific problem, the government asserts two broad objectives for s. 51(e): (1) to enhance civic responsibility and respect for the rule of law; and (2) to provide additional punishment or “enhance the general purposes of the criminal sanction”.  Vague and symbolic objectives, however,  make the justification analysis difficult.  The first objective could be asserted of virtually every criminal law and many non‑criminal measures.  Concerning the second objective, nothing in the record discloses precisely why Parliament felt that more punishment was required for this particular class of prisoner, or what additional objectives Parliament hoped to achieve by this punishment that were not accomplished by the sentences already imposed.  Nevertheless, rather than dismissing the government’s objectives outright,  prudence suggests that we  proceed to the proportionality inquiry.

Section 51(e) does not meet the proportionality test.  In particular, the government fails to establish a rational connection between s. 51(e)’s denial of the right to vote and its stated objectives.  With respect to the first objective of promoting civic responsibility and respect for the law, denying penitentiary inmates the right to vote is more likely to send messages that undermine respect for the law and democracy than messages that enhance those values.  The legitimacy of the law and the obligation to obey the law flow directly from the right of every citizen to vote.  To deny prisoners the right to vote is to lose an important means of teaching them democratic values and social responsibility.  The government’s novel political theory that would permit elected representatives to disenfranchise a segment of the population finds no place in a democracy built upon principles of inclusiveness, equality, and citizen participation.  That not all self‑proclaimed democracies adhere to this conclusion says little about what the Canadian vision of democracy embodied in the Charter  permits.  Moreover, the argument that only those who respect the law should participate in the political process cannot be accepted.  Denial of the right to vote on the basis of attributed moral unworthiness is inconsistent with the respect for the dignity of every person that lies at the heart of Canadian democracy and the Charter .  It also runs counter to the plain words of s. 3  of the Charter , its exclusion from the s. 33  override, and the idea that laws command obedience because they are made by those whose conduct they govern.

With respect to the second objective of imposing appropriate punishment,  the government offered no credible theory about why it should be allowed to deny a fundamental democratic right as a form of state punishment.  Denying the right to vote does not comply with the requirements for legitimate punishment — namely, that punishment must not be arbitrary and must serve a valid criminal law purpose.  Absence of arbitrariness requires that punishment be tailored to the acts and circumstances of the individual offender.  Section 51(e) qua  punishment bears little  relation to the offender’s particular crime.  As to a legitimate penal purpose,  neither the record nor common sense supports the claim that disenfranchisement deters crime or rehabilitates criminals.  By imposing a blanket punishment on all penitentiary inmates regardless of the particular crimes they committed, the harm they caused, or the normative character of their conduct, s. 51(e) does not meet the requirements of denunciatory, retributive punishment, and is not rationally connected to the government’s stated goal.

The impugned provision does not minimally impair the right to vote.  Section 51(e) is too broad, catching many people who, on the government’s own theory, should not be caught.  Section 51(e) cannot be saved by the mere fact that it is less restrictive than a blanket exclusion of all inmates from the franchise.

Lastly, the negative effects of denying citizens the right to vote would greatly outweigh the tenuous benefits that might ensue.  Denying prisoners the right to vote imposes negative costs on prisoners and on the penal system.  It removes a route to social development and undermines correctional law and policy directed towards rehabilitation and integration.  In light of the disproportionate number of Aboriginal people in penitentiaries, the negative effects of s. 51(e) upon prisoners have a disproportionate impact on Canada’s already disadvantaged Aboriginal population.

Since s. 51(e) unjustifiably infringes s. 3  of the Charter , it is unnecessary to consider the alternative argument that it infringes the equality guarantee of s. 15(1)  of the Charter .

Per L’Heureux‑Dubé, Gonthier, Major and  Bastarache JJ. (dissenting):   This case rests on philosophical, political and social considerations which are not capable of “scientific proof”.  It involves justifications for and against the limitation of the right to vote which are based upon axiomatic arguments of principle or value statements.  When faced with such justifications, this Court ought to turn to the text of s. 1  of the Charter  and to the basic principles which undergird both s. 1 and the relationship that provision has with the rights and freedoms protected within the Charter .  Particularly, s. 1  of the Charter  requires that this Court look to the fact that there may be different social or political philosophies upon which justifications for or against the limitations of rights may be based.  In such a context, where this Court is presented with competing social or political philosophies relating to the right to vote, it is not by merely approving or preferring one that the other is necessarily disproved or shown not to survive Charter  scrutiny.  If the social or political philosophy advanced by Parliament reasonably justifies a limitation of the right in the context of a free and democratic society, then it ought to be upheld as constitutional.  In the realm of competing social or political philosophies, reasonableness is the predominant s. 1 justification consideration. Section 1  of the Charter  does not constrain Parliament or authorize this Court to prioritize one reasonable social or political philosophy over reasonable others, but only empowers this Court to strike down those limitations which are not reasonable and which cannot be justified in a free and democratic society.  The decision before this Court is therefore not whether or not Parliament has made a proper policy decision, but whether the policy position chosen by Parliament is an acceptable choice amongst those permitted under the Charter .

Since this case is about evaluating choices regarding social or political philosophies and about shaping and giving practical application to values, especially values that may lie outside the Charter  but are of fundamental importance to Canadians, the “dialogue” between courts and Parliament is of particular importance.  The dialogue metaphor does not signal a lowering of the s. 1 justification standard. It simply suggests that when, after a full and rigorous s. 1 analysis, Parliament has satisfied the court that it has established a reasonable limit to a right that is demonstrably justified in a free and democratic society, the dialogue ends.  In this case, ‘dialogue’ has existed insofar as Parliament had been addressing, since well before the Federal Court of Appeal decision, an evaluation of the right to vote, and specifically, the issue of prisoner disenfranchisement.  This evaluation was obviously undertaken with the many cases concerning prisoner disenfranchisement that had occurred up to that point in mind.  In enacting s. 51 (e) of the Canada Elections Act and in providing a justification of that provision before the courts, Parliament has indicated that it has drawn a line.

In this case, while it has been conceded that s. 51 (e) of the Canada Elections Act infringes s. 3  of the Charter , the infringement is a reasonable limit that is demonstrably justified in a free and democratic society.  The objectives of s. 51( e )  are pressing and substantial.  Both objectives are based upon a reasonable and rational social or political philosophy.  The first objective, that of enhancing civic responsibility and respect for the rule of law, relates to the promotion of good citizenship.  The social rejection of serious crime reflects a moral line which safeguards the social contract and the rule of law and bolsters the importance of the nexus between individuals and the community.  The “promotion of civic responsibility” may be abstract or symbolic, but symbolic or abstract purposes can be valid of their own accord and must not be downplayed simply for the reason of their being symbolic.  The second objective is  the enhancement of the general purposes of the criminal sanction.  Section 51(e) clearly has a punitive aspect with a retributive function.  It is a valid objective for Parliament to develop appropriate sanctions and punishments for serious crime.  The disenfranchisement is a civil disability arising from the criminal conviction.

Section 51(e) meets the proportionality test.  First, the impugned legislation  is rationally connected to the objectives.  While a causal relationship between disenfranchising prisoners and the objectives is not empirically demonstrable, reason, logic and common sense, as well as extensive expert evidence, support a conclusion that there is a rational connection between disenfranchising offenders incarcerated for serious crimes and the objectives of promoting civic responsibility and the rule of law and the enhancement of the general objectives of the penal sanction.  With respect to the first objective, the removal of  the right to vote from serious incarcerated criminals does no injury to, but rather recognizes, their dignity.  Further, the disenfranchisement of serious criminal offenders serves to deliver a message to both the community and  the offenders themselves that serious criminal activity will not be tolerated by the community.  Society may choose to curtail temporarily the availability of the vote to serious criminals to insist that civic responsibility and respect for the rule of law, as goals worthy of pursuit, are prerequisites to democratic participation.  With respect to the second objective, the disenfranchisement is carefully tailored to apply to perpetrators of serious crimes, and  there is evidence in the record indicating that the denial of the right to vote is perceived as meaningful by the prisoners themselves and can therefore contribute to the rehabilitation of prisoners.  Lastly, many other democracies have, by virtue of choosing some form of prisoner disenfranchisement, also identified a connection between objectives similar to those advanced in the case at bar and the means of prisoner disenfranchisement.

Second, the impairment of the Charter  right is minimal.  Minimal impairment is about analyzing the line that has been drawn.  This analysis does not require the Crown to have adopted the absolutely least intrusive means for promoting the purpose, although it does require that the Crown prefer a significantly less intrusive means if it is of equal effectiveness.  Here, no less intrusive measure would be equally effective.  Only “serious offenders”, as determined by Parliament, are subject to disenfranchisement.  Since Parliament has drawn a two‑year cut off line which identifies which incarcerated offenders have committed serious enough crimes to warrant being deprived of the vote, any alternative line will not be of equal effectiveness.  Equal effectiveness is a dimension of the analysis that should not be underemphasized, as it relates directly to Parliament’s ability to pursue its legitimate objectives effectively.  Any other line insisted upon amounts to second‑guessing Parliament as to what constitutes a “serious” crime.  The provision is reasonably tailored insofar as disenfranchisement reflects the length of the sentence and actual incarceration, which, in turn, reflect the seriousness of the crime perpetrated and the intended progress towards the ultimate goals of rehabilitation and reintegration.  Section 51(e) is not arbitrary:  it is related directly to particular categories of conduct.  The two‑year cut off line also reflects several practical considerations.  Further, since this Court gave the impression that it was up to Parliament to do exactly this after the first Sauvé case was heard in 1993, there is a need for deference to Parliament in its drawing of a line.  The analysis of social and political philosophies and the accommodation of values in the context of the Charter  must be sensitive to the fact that there may be many possible reasonable and rational balances.  Line drawing, amongst a range of acceptable alternatives, is for Parliament, especially in this case where any alternate line would not be equally effective, in that the line drawn reflects Parliament’s identification of what amounts to serious criminal activity.

Third, when the objectives and the salutary effects are viewed in the totality of the context, they outweigh the temporary disenfranchisement of the serious criminal offender.  The enactment of the measure is itself a salutary effect.  The legislation intrinsically expresses societal values in relation to serious criminal behaviour and the right to vote in our society.  Value emerges from the signal or message that those who commit serious crimes will temporarily lose one aspect of the political equality of citizens.  Furthermore, the temporary disenfranchisement is perceived as meaningful by the offenders themselves and could have an ongoing positive rehabilitative effect.  The most obvious deleterious effect of s. 51(e) is the potential temporary loss of the vote.  This, however, must be considered in light of Parliament’s objectives, as illuminated by the totality of the context.  The  statistical data mentioned by the Federal Court of Appeal indicate that the provision catches serious and repeat offenders and that  most prisoners will only be deprived of participation in one election.  Because the duration of the disenfranchisement is directly related to the duration of incarceration, a serious criminal offender may never actually be denied the opportunity to vote if there is no election during the time he is incarcerated.  In light of the special context of this case — that the justification advanced by Parliament is rooted in a social or political philosophy that is not susceptible to proof in the traditional sense — deference is appropriate since the impugned provision raises questions of penal philosophy and policy.

Section 51(e) does not infringe s. 15(1)  of the Charter .  Even if it were presumed that the legislation draws a distinction based on personal characteristics, prisoners do not constitute a group protected by analogous or enumerated grounds under s. 15(1).  The fact of being incarcerated does not arise because of a stereotypical application of a presumed group characteristic.  The status of being a prisoner is brought about by the past commission of serious criminal offences, acts committed by the individual himself or herself.  The unifying group characteristic is past criminal behaviour.  The argument that imprisonment should be recognized as an analogous ground because of adverse effect or impact discrimination based on the fact that Aboriginal peoples make up a “disproportionate” percentage of prisoners must be rejected.  It is not plausible to say that the temporary disenfranchisement provision is in some way targeted at Aboriginal people.  The fact of incarceration does not necessarily arise due to any personal attribute such as race or ethnic origin and neither does it necessarily relate to social condition.  It hinges only upon the commission of serious criminal offences.

Cases Cited

By McLachlin C.J.

Referred to:  Sauvé v. Canada (Attorney General), [1993] 2 S.C.R. 438, aff’g (1992), 7 O.R. (3d) 481; R. v. Oakes, [1986] 1 S.C.R. 103; Hunter v. Southam Inc., [1984] 2 S.C.R. 145; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624; M. v. H., [1999] 2 S.C.R. 3;  Belczowski v. Canada, [1992] 2 F.C. 440; U.F.C.W., Local 1518 v. KMart Canada Ltd., [1999] 2 S.C.R. 1083; Thomson Newspapers Co. v. Canada (Attorney General), [1998] 1 S.C.R. 877; RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; R. v. Butler, [1992] 1 S.C.R. 452; R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2; Canadian Disability Rights Council v. Canada, [1988] 3 F.C. 622; Muldoon v. Canada, [1988] 3 F.C. 628; August v. Electoral Commission, 1999 (3) SALR 1; R. v. Smith, [1987] 1 S.C.R. 1045; R. v. M. (C.A.), [1996] 1 S.C.R. 500; R. v. Gladue, [1999] 1 S.C.R. 688.

By Gonthier J. (dissenting)

Sauvé v. Canada (Attorney General) (1988), 66 O.R. (2d) 234, rev’d (1992), 7 O.R. (3d) 481, aff’d [1993] 2 S.C.R. 438; R. v. Keegstra, [1990] 3 S.C.R. 697; R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2; R. v. Oakes, [1986] 1 S.C.R. 103; Reference re Provincial Electoral Boundaries (Sask.), [1991] 2 S.C.R. 158; United States of America v. Cotroni, [1989] 1 S.C.R. 1469; Thomson Newspapers Co. v. Canada (Attorney General), [1998] 1 S.C.R. 877; Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038; Harvey v. New Brunswick (Attorney General), [1996] 2 S.C.R. 876; R. v. Mills, [1999] 3 S.C.R. 668; Libman v. Quebec (Attorney General), [1997] 3 S.C.R. 569; RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; R. v. Butler, [1992] 1 S.C.R. 452; Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927; Reference Re Public Service Employee Relations Act (Alta.), [1987] 1 S.C.R. 313; Canada v. Schmidt, [1987] 1 S.C.R. 500; Canada (Auditor General) v. Canada (Minister of Energy, Mines and Resources), [1989] 2 S.C.R. 49; Vriend v. Alberta, [1998] 1 S.C.R. 493; Re Jolivet and The Queen (1983), 1 D.L.R. (4th) 604; Gould v. Canada (Attorney General), [1984] 2 S.C.R. 124, aff’g [1984] 1 F.C. 1133, rev’g [1984] 1 F.C. 1119; Lévesque v. Canada (Attorney General), [1986] 2 F.C. 287; Badger v. Attorney-General of Manitoba (1986), 30 D.L.R. (4th) 108, aff’d (1986), 32 D.L.R. (4th) 310; Badger v. Canada (Attorney General) (1988), 55 Man. R. (2d) 211, rev’d (1988), 55 D.L.R. (4th) 177, leave to appeal refused, [1989] 1 S.C.R. v; Belczowski v. Canada, [1991] 3 F.C. 151, aff’d [1992] 2 F.C. 440, aff’d [1993] 2 S.C.R. 438; Little Sisters Book and Art Emporium v. Canada (Minister of Justice), [2000] 2 S.C.R. 1120, 2000 SCC 69; Pearson v. Secretary of State for the Home Department, [2001] E.W.J. No. 1566 (QL); Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721; Reference re Secession of Quebec, [1998] 2 S.C.R. 217; Operation Dismantle Inc. v. The Queen, [1985] 1 S.C.R. 441; Haig v. Canada, [1993] 2 S.C.R. 995; Driskell v. Manitoba (Attorney General), [1999] 11 W.W.R. 615; Byatt v. Dykema (1998), 158 D.L.R. (4th) 644; Richardson v. Ramirez, 418 U.S. 24 (1974); X. v. Netherlands, Application No. 6573/74, December 19, 1974, D.R. 1, p. 87; H. v. Netherlands, Application No. 9914/82, July 4, 1983, D.R. 33, p. 242; Holland v. Ireland, Application No. 24827/94, April 14, 1998, D.R. 93-A, p. 15; Eur. Court. H. R., Mathieu-Mohin and Clerfayt case, judgment of 2 March 1987, Series A No. 113; Lavigne v. Ontario Public Service Employees Union, [1991] 2 S.C.R. 211; Ross v. New Brunswick School District No. 15, [1996] 1 S.C.R. 825; R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 713; McKinney v. University of Guelph, [1990] 3 S.C.R. 229; Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835; R. v. Goltz, [1991] 3 S.C.R. 485; R. v. Guiller (1985), 48 C.R. (3d) 226; R. v. Luxton, [1990] 2 S.C.R. 711; R. v. Smith, [1987] 1 S.C.R. 1045; Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497; M. v. H., [1999] 2 S.C.R. 3; Jackson v. Joyceville Penitentiary, [1990] 3 F.C. 55; McKinnon v. M.N.R., 91 D.T.C. 1002; Armstrong v. R., [1996] 1 C.T.C. 2745; Mulligan v. R., [1997] 2 C.T.C. 2062; Wells v. R., [1998] 1 C.T.C. 2118; Olson v. Canada, [1996] 2 F.C. 168, leave to appeal refused, [1997] 3 S.C.R. xii; Alcorn v. Canada (Commissioner of Corrections) (1999), 163 F.T.R. 1, aff’d (2002), 95 C.R.R. (2d) 326, 2002 FCA 154; Chiarelli v. Canada (Minister of Employment and Immigration), [1992] 1 S.C.R. 711, rev’g [1990] 2 F.C. 299; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624; R. v. Gladue, [1999] 1 S.C.R. 688.

Statutes and Regulations Cited

Canada Elections Act, R.S.C. 1985, c. E-2, ss. 51(e) [rep. & sub. 1993, c. 19, s. 23(2)], 51.1 [ad. idem, s. 24].

Canada Elections Act , S.C. 2000, c. 9 , s. 4(c).

Canadian Charter of Rights and Freedoms , preamble, ss. 1, 3, 6(1), 11, 12, 15(1), 33.

Constitution Act, 1982 , s. 52 .

Convention for the Protection of Human Rights and Fundamental Freedoms, 213 U.N.T.S. 221.

Criminal Code , R.S.C. 1985, c. C-46 , ss. 718.2 (e), 742.1 , 743.1 , 743.6 .

Election Act, R.S.A. 2000, c. E-1, s. 45(c).

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

Election Act, R.S.Q., c. E-3.3, s. 273.

Election Act, S.P.E.I. 1996, c. 12.

Election Act, 1996, S.S. 1996, c. E-6.01, s. 17.

Election Statute Law Amendment Act, 1998, S.O. 1998, c. 9, s. 13.

Elections Act, R.S.M. 1987, c. E30, s. 31 [rep. & sub. 1998, c. 4, s. 21].

Elections Act, R.S.N.B. 1973, c. E-3, s. 43(2)(e).

Elections Act, R.S.N.S. 1989, c. 140, s. 29(d) [am. 2001, c. 43, s. 13].

Elections Act, R.S.N.W.T. 1988, c. E-2, s. 27(3) [am. 1995, c. 14, s. 6].

Elections Act, R.S.Y. 1986, c. 48, s. 5(d).

Elections Act, 1991, S.N. 1992, c. E-3.1.

European Assembly Elections Act 1978 (U.K.), 1978, c. 10, Sch. 1, s. 2(1).

Human Rights Act 1998 (U.K.), 1998, c. 42.

International Covenant on Civil and Political Rights, 999 U.N.T.S. 171, Art. 25.

Protocol to the Convention for the Protection of Human Rights and Fundamental Freedoms, Eur. T.S. No. 9, Art. 3.

Representation of the People Act 1983 (U.K.), 1983, c. 2,  s. 3(1) [am. 1985, c. 50, Sch. 4, s. 1; am. 2000, c. 2, s. 2].

United States Constitution, Article 1, Fourteenth Amendment, Fifteenth Amendment, Seventeenth Amendment, Nineteenth Amendment, Twenty-Fourth Amendment, Twenty-Sixth Amendment.

Authors Cited

Canada.  House of Commons.  House of Commons Debates, vol. XIV, 3rd Sess., 34th Parl., April 2, 1993, pp. 18015-021. 

Canada.  Royal Commission on Electoral Reform and Party Financing.  Final Report of the Royal Commission on Electoral Reform and Party Financing — Reforming Electoral Democracy, vol. 1.  Ottawa:  The Commission, 1991.

Canada.  Senate and House of Commons.  Special Joint Committee on the Constitution of Canada.  Minutes of Proceedings and Evidence, Issue No. 43, 1st Sess., 32nd Parl., January 22, 1981, pp. 43:79-43:90.

Concise Oxford Dictionary of Current English, 9th ed.  Oxford:  Clarendon Press, 1995, “moral”.

Dworkin, Ronald. Taking Rights Seriously.  Cambridge: Harvard University Press, 1977.

George, Robert P.  Making Men Moral.  Oxford:  Clarendon Press, 1993.

Grimm, Dieter.  “Human Rights and Judicial Review in Germany”, in David M. Beatty, ed., Human Rights and Judicial Review:  A Comparative Perspective.  Dordrecht:  Martinus Nijhoff Publishers, 1994, 267.

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

Hogg, Peter W.  Constitutional Law of Canada, vol. 2, loose-leaf ed.  Scarborough, Ont.:  Carswell, 1992 (updated 2001, release 1).

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

Macklem, P., et al.  Canadian Constitutional Law, 2nd ed. Toronto: Emond Montgomery Publications, 1997.

Mill, John Stuart.  “Thoughts on Parliamentary Reform”, in J. M. Robson, ed., Essays on Politics and Society, vol. XIX.  Toronto:  University of Toronto Press, 1977, 311.

Penal Reform International.  Making Standards Work — An International Handbook on Good Prison Practice.  The Hague: Penal Reform International, 1995.

Raz, Joseph.  The Morality of Freedom.  Oxford:  Clarendon Press, 1986.

Tribe, Laurence H.  American Constitutional Law, 2nd ed.  Mineola, New York:  Foundation Press, 1988.

Tribe, Laurence H.  “The Disenfranchisement of Ex‑Felons:  Citizenship, Criminality, and ‘The Purity of the Ballot Box’” (1989), 102 Harv. L. Rev. 1300.

United Nations.  Human Rights Committee.  “General Comment Adopted by the Human Rights Committee under Article 40, Paragraph 4, of the International Covenant on Civil and Political Rights”, General Comment No. 25 (57), Annex V, CCPR/C/21, Rev. 1, Add. 7, August 27, 1996.

APPEAL from a judgment of the Federal Court of Appeal, [2000] 2 F.C. 117, 180 D.L.R. (4th) 385, 248 N.R. 267, 29 C.R. (5th) 242, 69 C.R.R. (2d) 106, [1999] F.C.J. No. 1577 (QL), allowing the respondents’ appeal and dismissing the appellants’ cross-appeal from a decision of the Trial Division, [1996] 1 F.C. 857, 106 F.T.R. 241, 132 D.L.R. (4th) 136, [1995] F.C.J. No. 1735 (QL).  Appeal allowed, L’Heureux‑Dubé, Gonthier, Major and  Bastarache JJ. dissenting.

Fergus J. O'Connor, for the appellant Richard Sauvé.

Arne Peltz, for the appellants Sheldon McCorrister, Lloyd Knezacek, Clair Woodhouse, Aaron Spence, Serge Bélanger, Emile A. Bear and Randy Opoonechaw.

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

Thomas W. Wakeling and Gerald D. Chipeur, for the intervener the Attorney General for Alberta.

Heather S. Leonoff, Q.C., for the intervener the Attorney General of Manitoba.

Allan Manson and Elizabeth Thomas, for the interveners the Canadian Association of Elizabeth Fry Societies and the John Howard Society of Canada.

John W. Conroy, Q.C., for the intervener the British Columbia Civil Liberties Association.

Kent Roach and Brian Eyolfson, for the intervener the Aboriginal Legal Services of Toronto Inc.

Sylvain Lussier, for the intervener the Canadian Bar Association.

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

1                                   The Chief Justice — The right of every citizen to vote, guaranteed by s. 3  of the Canadian Charter of Rights and Freedoms , lies at the heart of Canadian democracy.  The law at stake in this appeal denies the right to vote to a certain class of people — those serving sentences of two years or more in a correctional institution.  The question is whether the government has established that this denial of the right to vote is allowed under s. 1  of the Charter  as a “reasonable limi[t] . . . demonstrably justified in a free and democratic society”.  I conclude that it is not.  The right to vote, which lies at the heart of Canadian democracy, can only be trammeled for good reason.  Here, the reasons offered do not suffice.

I.  Statutory Provisions

2                                   The predecessor to s. 51(e) of the Canada Elections Act, R.S.C. 1985, c. E-2, prohibited all prison inmates from voting in federal elections, regardless of the length of their sentences.  This section was held unconstitutional as an unjustified denial of the right to vote guaranteed by s. 3  of the Charter : Sauvé v. Canada (Attorney General), [1993] 2 S.C.R. 438.  Parliament responded to this litigation by replacing this section with a new s. 51(e) (S.C. 1993, c. 19, s. 23), which denies the right to vote to all inmates serving sentences of two years or more.  Section 51(e), which is now continued in substantially the same form at s. 4(c) of the Act (S.C. 2000, c. 9 ), and the relevant Charter  provisions are set out below.

Canada Elections Act, R.S.C. 1985, c. E-2

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;

Canadian Charter of Rights and Freedoms 

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.

                                                                   . . .

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.

                                                                   . . .

15. (1)  Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

II.  Judgments

A.  Federal Court, Trial Division, [1996] 1 F.C. 857

3                                   The trial judge, Wetston J., held that s. 51 (e) of the Canada Elections Act violated the Charter  guarantee of the right to vote without being demonstrably justified, and was therefore void.  Although he found that the government’s objectives were pressing and substantial, he concluded that the denial of voting rights to all inmates serving a sentence of two years or longer was overbroad and failed the minimal impairment test.  In addition, he found at p. 913 that denying the right to vote “hinder[ed] the rehabilitation of offenders and their successful reintegration into the community”.  The negative consequences of the challenged provision were thus disproportionate to any benefits it might produce. 

B.  Federal Court of Appeal, [2000] 2 F.C. 117

4                                   The majority of the Federal Court of Appeal, per Linden J.A., reversed the trial judge and upheld the denial of voting rights, holding that Parliament’s role in maintaining and enhancing the integrity of the electoral process and in exercising the criminal law power both warranted deference.  The denial of the right to vote at issue fell within a reasonable range of alternatives open to Parliament to achieve its objectives and was not overbroad or disproportionate.  Desjardins J.A., applying the “stringent formulation of the Oakes test”, emphasized the absence of evidence of benefits flowing from the denial and would have dismissed the appeal.

III.  Issues

5                                   1.    Does s. 51 (e) of the Canada Elections Act infringe the guarantee of the right of all citizens to vote under s. 3  of the Charter  and if so, is the infringement justified under s. 1  of the Charter ?

2.                                 Does s. 51 (e) of the Canada Elections Act infringe the equality guarantee of s. 15(1)  of the Charter  and if so, is the infringement justified under s. 1  of the Charter ?

IV.  Analysis

6                                   The respondents concede that the voting restriction at issue violates s. 3  of the Charter .  The restriction is thus invalid unless demonstrably justified under s. 1.  I shall therefore proceed directly to the s. 1 analysis. 

A.  The Approach to Section 1 Justification

7                                   To justify the infringement of a Charter  right, the government must show that the infringement achieves a constitutionally valid purpose or objective, and that the chosen means are reasonable and demonstrably justified: R. v. Oakes, [1986] 1 S.C.R. 103.  This two-part inquiry — the legitimacy of the objective and the proportionality of the means — ensures that a reviewing court examine rigorously all aspects of justification.  Throughout the justification process, the government bears the burden of proving a valid objective and showing that the rights violation is warranted — that is, that it is rationally connected, causes minimal impairment, and is proportionate to the benefit achieved.

8                                   My colleague Justice Gonthier proposes a deferential approach to infringement and justification.  He argues that there is no reason to accord special importance to the right to vote, and that we should thus defer to Parliament’s choice among a range of reasonable alternatives.  He further argues that in justifying limits on the right to vote under s. 1, we owe deference to Parliament because we are dealing with “philosophical, political and social considerations”, because of the abstract and symbolic nature of the government’s stated goals, and because the law at issue represents a step in a dialogue between Parliament and the courts.

9                                    I must, with respect, demur.  The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside.  Limits on it require not deference, but careful examination.  This is not a matter of substituting the Court’s philosophical preference for that of the legislature, but of ensuring that the legislature’s proffered justification is supported by logic and common sense.

10                               The Charter distinguishes between two separate issues: whether a right has been infringed, and whether the limitation is justified.  The complainant bears the burden of showing the infringement of a right (the first step), at which point the burden shifts to the government to justify the limit as a reasonable limit under s. 1 (the second step).  These are distinct processes with different burdens.  Insulating a rights restriction from scrutiny by labeling it a matter of social philosophy, as the government attempts to do, reverses the constitutionally imposed burden of justification.  It removes the infringement from our radar screen, instead of enabling us to zero in on it to decide whether it is demonstrably justified as required by the Charter .

11                               At the first stage, which involves defining the right, we must follow this Court’s consistent view that rights shall be defined broadly and liberally: Hunter v. Southam Inc.,  [1984] 2 S.C.R. 145, at p. 156; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 344; Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624, at para. 53. A broad and purposive interpretation of the right is particularly critical in the case of the right to vote.  The framers of the Charter  signaled the special importance of this right not only by its broad, untrammeled language, but by exempting it from legislative override under s. 33 ’s notwithstanding clause.  I conclude that s. 3 must be construed as it reads, and its ambit should not be limited by countervailing collective concerns, as the government appears to argue.  These concerns are for the government to raise under s. 1 in justifying the limits it has imposed on the right.

12                               At the s. 1 stage, the government argues that denying the right to vote to penitentiary inmates is a matter of social and political philosophy, requiring deference.  Again, I cannot agree.  This Court has repeatedly held that the “general claim that the infringement of a right is justified under s. 1” does not warrant deference to Parliament: M. v. H., [1999] 2 S.C.R. 3, at para. 78, per Iacobucci J.  Section 1 does not create a presumption of constitutionality for limits on rights; rather, it requires the state to justify such limitations. 

13                               The core democratic rights of Canadians do not fall within a “range of acceptable alternatives” among which Parliament may pick and choose at its discretion.  Deference may be appropriate on a decision involving competing social and political policies. It is not appropriate, however, on a decision to limit fundamental rights.  This case is not merely a competition between competing social philosophies.  It represents a conflict between the right of citizens to vote — one of the most fundamental rights guaranteed by the Charter  — and Parliament’s denial of that right.  Public debate on an issue does not transform it into a matter of “social philosophy”,  shielding it from full judicial scrutiny.  It is for the courts, unaffected by the shifting winds of public opinion and electoral interests, to safeguard the right to vote guaranteed by s. 3  of the Charter .

14                               Charter  rights are not a matter of privilege or merit, but a function of membership in the Canadian polity that cannot lightly be cast aside.  This is manifestly true of the right to vote, the cornerstone of democracy, exempt from the incursion permitted on other rights through s. 33  override.  Thus, courts considering denials of voting rights have applied a stringent justification standard: Sauvé v. Canada (Attorney General) (1992), 7 O.R. (3d) 481 (C.A.) (“Sauvé No. 1”), and Belczowski v. Canada, [1992] 2 F.C. 440 (C.A.). 

15                               The Charter charges courts with upholding and maintaining an inclusive, participatory democratic framework within which citizens can explore and pursue different conceptions of the good.  While a posture of judicial deference to legislative decisions about social policy may be appropriate in some cases, the legislation at issue does not fall into this category.  To the contrary, it is precisely when legislative choices threaten to undermine the foundations of the participatory democracy guaranteed by the Charter  that courts must be vigilant in fulfilling their constitutional duty to protect the integrity of this system.

16                               Nor can I concur in the argument that the philosophically based or symbolic nature of the government’s objectives in itself commands deference.  To the contrary, this Court has held that broad, symbolic objectives are problematic, as I discuss below: see U.F.C.W., Local 1518 v. KMart Canada Ltd., [1999] 2 S.C.R. 1083, at para. 59, per Cory J.; Thomson Newspapers Co. v. Canada (Attorney General), [1998] 1 S.C.R. 877, at para. 87, per Bastarache J.; RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199, at paras. 143-44, per McLachlin J. (as she then was).  Parliament cannot use lofty objectives to shield legislation from Charter  scrutiny.  Section 1 requires valid objectives and proportionality.

17                               Finally, the fact that the challenged denial of the right to vote followed judicial rejection of an even more comprehensive denial, does not mean that the Court should defer to Parliament as part of a “dialogue”.  Parliament must ensure that whatever law it passes, at whatever stage of the process, conforms to the Constitution.  The healthy and important promotion of a dialogue between the legislature and the courts should not be debased to a rule of “if at first you don’t succeed, try, try again”.

18                               While deference to the legislature is not appropriate in this case, legislative justification does not require empirical proof in a scientific sense.  While some matters can be proved with empirical or mathematical precision, others, involving philosophical, political and social considerations, cannot.  In this case, it is enough that the justification be convincing, in the sense that it is sufficient to satisfy the reasonable person looking at all the evidence and relevant considerations, that the state is justified in infringing the right at stake to the degree it has: see RJR-MacDonald, supra, at para. 154, per McLachlin J.; R. v. Butler, [1992] 1 S.C.R. 452, at pp. 502-3, per Sopinka J.  What is required is “rational, reasoned defensibility”: RJR-MacDonald, at para. 127.  Common sense and inferential reasoning may supplement the evidence: R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2,  at para. 78, per McLachlin C.J.  However, one must be wary of stereotypes cloaked as common sense, and of substituting deference for the reasoned demonstration required by s. 1.

19                               Keeping in mind these basic principles of Charter  review, I approach the familiar stages of the Oakes test.  I conclude that the government’s stated objectives of promoting civic responsibility and respect for the law and imposing appropriate punishment, while problematically vague, are capable in principle of justifying limitations on Charter  rights. However, the government fails to establish proportionality, principally for want of a rational connection between denying the vote to penitentiary inmates and its stated goals.

B.  The Government’s Objectives

20                               The objectives’ analysis entails a two-step inquiry.  First, we must ask what the objectives are of denying penitentiary inmates the right to vote.  This involves interpretation and construction, and calls for a contextual approach: Thomson Newspapers, supra, at para. 87.  Second, we must evaluate whether the objectives as found are capable of justifying limitations on Charter  rights.  The objectives must not be “trivial”, and they must not be “discordant with the principles integral to a free and democratic society”: Oakes, supra, at p. 138.  To borrow from the language of German constitutional law, there must be a constitutionally valid reason for infringing a right: see D. Grimm, “Human Rights and Judicial Review in Germany”, in D. M. Beatty, ed., Human Rights and Judicial Review: A Comparative Perspective (1994), 267, at p. 275.  Because s. 1 serves first and foremost to protect rights, the range of constitutionally valid objectives is not unlimited.  For example, the protection of competing rights might be a valid objective.  However, a simple majoritarian political preference for abolishing a right altogether would not be a constitutionally valid objective. 

21                               Section 51(e) denying penitentiary inmates the right to vote was not directed at a specific problem or concern. Prisoners have long voted, here and abroad, in a variety of situations without apparent adverse effects to the political process, the prison population, or society as a whole.  In the absence of a specific problem, the government asserts two broad objectives as the reason for this denial of the right to vote: (1) to enhance civic responsibility and respect for the rule of law; and (2) to provide additional punishment, or “enhanc[e] the general purposes of the criminal sanction”.  The record leaves in doubt how much these goals actually motivated Parliament; the Parliamentary debates offer more fulmination than illumination.   However, on the basis of “some glimmer of light”, the trial judge at p. 878 concluded that they could be advanced as objectives of the denial.  I am content to proceed on this basis.

22                               This leaves the question of whether the objectives of enhancing respect for law and appropriate punishment are constitutionally valid and sufficiently significant to warrant a rights violation.  Vague and symbolic objectives such as these almost guarantee a positive answer to this question. Who can argue that respect for the law is not pressing?  Who can argue that proper sentences are not important?  Who can argue that either of these goals, taken at face value, contradicts democratic principles?  However, precisely because they leave so little room for argument, vague and symbolic objectives make the justification analysis more difficult.  Their terms carry many meanings, yet tell us little about why the limitation on the right is necessary, and what it is expected to achieve in concrete terms.  The broader and more abstract the objective, the more susceptible it is to different meanings in different contexts, and hence to distortion and manipulation.  One articulation of the objective might inflate the importance of the objective; another might make the legislative measure appear more narrowly tailored.  The Court is left to sort the matter out. 

23                               At the end of the day, people should not be left guessing about why their Charter  rights have been infringed.  Demonstrable justification requires that the objective clearly reveal the harm that the government hopes to remedy, and that this objective remain constant throughout the justification process.  As this Court has stated, the objective “must be accurately and precisely defined so as to provide a clear framework for evaluating its importance, and to assess the precision with which the means have been crafted to fulfil that objective”: per Cory J. in U.F.C.W., Local 1518, supra, at para. 59; see also Thomson Newspapers, supra, at para. 96; RJR-MacDonald, supra, at para. 144.  A court faced with vague objectives may well conclude, as did Arbour J.A. (as she then was) in Sauvé No. 1, supra, at p. 487, that “the highly symbolic and abstract nature of th[e] objective . . . detracts from its importance as a justification for the violation of a constitutionally protected right”.  If Parliament can infringe a crucial right such as the right to vote simply by offering symbolic and abstract reasons, judicial review either becomes vacuously constrained or reduces to a contest of “our symbols are better than your symbols”.  Neither outcome is compatible with the vigorous justification analysis required by the Charter .

24                               The rhetorical nature of the government objectives advanced in this case renders them suspect.  The first objective, enhancing civic responsibility and respect for the law, could be asserted of virtually every criminal law and many non-criminal measures.  Respect for law is undeniably important.  But the simple statement of this value lacks the context necessary to assist us in determining whether the infringement at issue is demonstrably justifiable in a free and democratic society.  To establish justification, one needs to know what problem the government is targeting, and why it is so pressing and important that it warrants limiting a Charter  right.  Without this, it is difficult if not impossible to weigh whether the infringement of the right is justifiable or proportionate.

25                               The second objective — to impose additional punishment on people serving penitentiary sentences — is less vague than the first.  Still, problems with vagueness remain.  The record does not disclose precisely why Parliament felt that more punishment was required for this particular class of prisoner, or what additional objectives Parliament hoped to achieve by this punishment that were not accomplished by the sentences already imposed.  This makes it difficult to assess whether the objective is important enough to justify an additional rights infringement.                           

26                               Quite simply, the government has failed to identify particular problems that require denying the right to vote, making it hard to say that the denial is directed at a pressing and substantial purpose.  Nevertheless, despite the abstract nature of the government’s objectives and the rather thin basis upon which they rest, prudence suggests that we proceed to the proportionality analysis, rather than dismissing the government’s objectives outright.  The proportionality inquiry allows us to determine whether the government’s asserted objectives are in fact capable of justifying its denial of the right to vote.  At that stage, as we shall see, the difficulties inherent in the government’s stated objectives become manifest.

C.  Proportionality

27                               At this stage the government must show that the denial of the right to vote will promote the asserted objectives (the rational connection test); that the denial does not go further than reasonably necessary to achieve its objectives (the minimal impairment test); and that the overall benefits of the measure outweigh its negative impact (the proportionate effect test).  As will be seen, the vagueness of the government’s justificatory goals coupled with the centrality of the right to vote to Canadian democracy, the rule of law, and legitimate sentencing, make the government’s task difficult indeed.

1.  Rational Connection

28                               Will denying the right to vote to penitentiary inmates enhance respect for the law and impose legitimate punishment?  The government must show that this is likely, either  by evidence or in reason and logic: RJR-MacDonald, supra, at para. 153.

29                               The government advances three theories to demonstrate rational connection between its limitation and the objective of enhancing respect for law. First, it submits that depriving penitentiary inmates of the vote sends an “educative message” about the importance of respect for the law to inmates and to the citizenry at large.  Second, it asserts that allowing penitentiary inmates to vote “demeans” the political system.   Finally, it takes the position that disenfranchisement is a legitimate form of punishment, regardless of the specific nature of the offence or the circumstances of the individual offender.  In my respectful view, none of these claims succeed.

30                               The first asserted connector with enhancing respect for the law is the “educative message” or “moral statement” theory.  The problem here, quite simply, is that denying penitentiary inmates the right to vote is bad pedagogy.  It misrepresents the nature of our rights and obligations under the law, and it communicates a message more likely to harm than to help respect for the law. 

31                               Denying penitentiary inmates the right to vote misrepresents the nature of our rights and obligations under the law and consequently undermines them.   In a democracy such as ours, the power of lawmakers flows from the voting citizens, and lawmakers act as the citizens’ proxies.  This delegation from voters to legislators gives the law its legitimacy or force.  Correlatively, the obligation to obey the law flows from the fact that the law is made by and on behalf of the citizens.  In sum, the legitimacy of the law and the obligation to obey the law flow directly from the right of every citizen to vote.  As a practical matter, we require all within our country’s boundaries to obey its laws, whether or not they vote.  But this does not negate the vital symbolic, theoretical and practical connection between having a voice in making the law and being obliged to obey it.  This connection, inherited from social contract theory and enshrined in the Charter , stands at the heart of our system of constitutional democracy.

32                               The government gets this connection exactly backwards when it attempts to argue that depriving people of a voice in government teaches them to obey the law.  The “educative message” that the government purports to send by disenfranchising inmates is both anti-democratic and internally self-contradictory.   Denying a citizen the right to vote denies the basis of democratic legitimacy.  It says that delegates elected by the citizens can then bar those very citizens, or a portion of them, from participating in future elections.  But if we accept that governmental power in a democracy flows from the citizens, it is difficult to see how that power can legitimately be used to disenfranchise the very citizens from whom the government’s power flows.

33                               Reflecting this truth, the history of democracy is the history of progressive enfranchisement.  The universal franchise has become, at this point in time, an essential part of democracy.  From the notion that only a few meritorious people could vote (expressed in terms like class, property and gender), there gradually evolved the modern precept that all citizens are entitled to vote as members of a self-governing citizenry.  Canada’s steady march to universal suffrage culminated in 1982, with our adoption of a constitutional guarantee of the right of all citizens to vote in s. 3  of the Charter .  As Arbour J.A.  observed in Sauvé No. 1, supra, at p. 487:

. . . 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.

Under s. 3  of the Charter , the final vestiges of the old policy of selective voting have fallen, including the exclusion of persons with a “mental disease” and federally appointed judges: see Canadian Disability Rights Council v. Canada, [1988] 3 F.C. 622 (T.D.); and Muldoon v. Canada, [1988] 3 F.C. 628 (T.D.).  The disenfranchisement of inmates takes us backwards in time and retrenches our democratic entitlements.  

34                               The right of all citizens to vote, regardless of virtue or mental ability or other distinguishing features, underpins the legitimacy of Canadian democracy and Parliament’s claim to power.  A government that restricts the franchise to a select portion of citizens is a government that weakens its ability to function as the legitimate representative of the excluded citizens, jeopardizes its claim to representative democracy, and erodes the basis of its right to convict and punish law-breakers.

35                               More broadly, denying citizens the right to vote runs counter to our constitutional commitment to the inherent worth and dignity of every individual.  As the South African Constitutional Court said in August v. Electoral Commission, 1999 (3) SALR 1, at para. 17, “[t]he vote of each and every citizen is a badge of dignity and of personhood.  Quite literally, it says that everybody counts.”  The fact that the disenfranchisement law at issue applies to a discrete group of persons should make us more, not less, wary of its potential to violate the principles of equal rights and equal membership embodied in and protected by the Charter .

36                               In recognition of the seminal importance of the right to vote in the constellation of rights, the framers of the Charter  accorded it special protections.  Unlike other rights, the right of every citizen to vote cannot be suspended under the “notwithstanding clause”.  As Arbour J.A. said in Sauvé No. 1, supra, at p. 486:

It is indeed significant that s. 3  of the Charter  is immune from the notwithstanding clause contained in s. 33, which permits Parliament and the legislatures to enact legislation which would otherwise violate the Charter . It confirms that the right to vote must be protected against those who have the capacity, and often the interest, to limit the franchise.  Unpopular minorities may seek redress against an infringement of their rights in the courts.  But like everybody else, they can only seek redress against a dismissal of their political point of view at the polls.

37                               The government’s vague appeal to “civic responsibility” is unhelpful, as is the attempt to lump inmate disenfranchisement together with legitimate voting regulations in support of the government’s position.  The analogy between youth voting restrictions and inmate disenfranchisement breaks down because the type of judgment Parliament is making in the two scenarios is very different. In the first case, Parliament is making a decision based on the experiential situation of all citizens when they are young. It is not saying that the excluded class is unworthy to vote, but regulating a modality of the universal franchise.  In the second case, the government is making a decision that some people, whatever their abilities, are not morally worthy to vote — that they do not  “deserve” to be considered members of the community and hence may be deprived of the most basic of their constitutional rights. But this is not the lawmakers’ decision to make.  The Charter makes this decision for us by guaranteeing the right of “every citizen” to vote and by expressly placing prisoners under the protective umbrella of the Charter  through constitutional limits on punishment. The Charter emphatically says that prisoners are protected citizens, and short of a constitutional amendment, lawmakers cannot change this.

38                               The theoretical and constitutional links between the right to vote and respect for the rule of law are reflected in the practical realities of the prison population and the need to bolster, rather than to undermine, the feeling of connection between prisoners and society as a whole.  The government argues that disenfranchisement will “educate” and rehabilitate inmates.  However, disenfranchisement is more likely to become a self-fulfilling prophecy than a spur to reintegration.  Depriving at-risk individuals of their sense of collective identity and membership in the community is unlikely to instill a sense of responsibility and community identity, while the right to participate in voting helps teach democratic values and social responsibility (testimony of Professor Jackson, appellants’ record at pp. 2001-2).  As J. S. Mill wrote:

To take an active interest in politics is, in modern times, the first thing which elevates the mind to large interests and contemplations; the first step out of the narrow bounds of individual and family selfishness, the first opening in the contracted round of daily occupations. . . . The possession and the exercise of political, and among others of electoral, rights, is one of the chief instruments both of moral and of intellectual training for the popular mind . . . .

(J. S. Mill, “Thoughts on Parliamentary Reform” (1859), in J. M. Robson, ed., Essays on Politics and Society, vol. XIX, 1977, 311, at pp. 322-23)

To deny prisoners the right to vote is to lose an important means of teaching them democratic values and social responsibility.

39                               Even if these difficulties could be overcome, it is not apparent that denying penitentiary inmates the right to vote actually sends the intended message to prisoners, or to the rest of society.  People may be sentenced to imprisonment for two years or more for a wide variety of crimes, ranging from motor vehicle and regulatory offences to the most serious cases of murder.  The variety of offences and offenders covered by the prohibition suggests that the educative message is, at best, a mixed and diffuse one.

40                               It is a message sullied, moreover, by negative and unacceptable messages likely to undermine civic responsibility and respect for the rule of law.  Denying citizen law-breakers the right to vote sends the message that those who commit serious breaches are no longer valued as members of the community, but instead are temporary outcasts from our system of rights and democracy.  More profoundly, it sends the unacceptable message that democratic values are less important than punitive measures ostensibly designed to promote order.  If modern democratic history has one lesson to teach it is this: enforced conformity to the law should not come at the cost of our core democratic values.

41                               I conclude that denying penitentiary inmates the right to vote is more likely to send messages that undermine respect for the law and democracy than messages that enhance those values.  The government’s novel political theory that would permit elected representatives to disenfranchise a segment of the population finds no place in a democracy built upon principles of inclusiveness, equality, and citizen participation.  That not all self-proclaimed democracies adhere to this conclusion says little about what the Canadian vision of democracy embodied in the Charter  permits.  Punitive disenfranchisement of inmates does not send the “educative message” that the government claims; to the contrary, it undermines this message and is incompatible with the basic tenets of participatory democracy contained in and guaranteed by the Charter .

42                               The government also argues that denying penitentiary inmates the vote will enhance respect for law because allowing people who flaunt the law to vote demeans the political system.  The same untenable premises we have been discussing resurface here — that voting is a privilege the government can suspend and that the commission of a serious crime signals that the offender has chosen to “opt out” of community membership.  But beyond this, the argument that only those who respect the law should participate in the political process is a variant on the age-old unworthiness rationale for denying the vote. 

43                               The idea that certain classes of people are not morally fit or morally worthy to vote and to participate in the law-making process is ancient and obsolete.  Edward III pronounced that citizens who committed serious crimes suffered “civil death”, by which a convicted felon was deemed to forfeit all civil rights.  Until recently, large classes of people, prisoners among them, were excluded from the franchise.  The assumption that they were not fit or “worthy” of voting — whether by reason of class, race, gender or conduct — played a large role in this exclusion.  We should reject the retrograde notion that “worthiness” qualifications for voters may be logically viewed as enhancing the political process and respect for the rule of law.  As Arbour J.A. stated in Sauvé No. 1, supra, at p. 487, since the adoption of s. 3  of the Charter , it is doubtful “that anyone could now be deprived of the vote on the basis . . . that he or she was not decent or responsible”.

44                               Denial of the right to vote on the basis of attributed moral unworthiness is inconsistent with the respect for the dignity of every person that lies at the heart of Canadian democracy and the Charter : compare August, supra.  It also runs counter to the plain words of s. 3, its exclusion from the s. 33  override, and the idea that laws command obedience because they are made by those whose conduct they govern.  For all these reasons, it must, at this stage of our history, be rejected.  

45                               This brings us to the government’s final argument for rational connection — that disenfranchisement is a legitimate weapon in the state’s punitive arsenal against the individual lawbreaker.  Again, the argument cannot succeed.  The first reason is that using the denial of rights as punishment is suspect.  The second reason is that denying the right to vote does not comply with the requirements for legitimate punishment established by our jurisprudence. 

46                               The argument, stripped of rhetoric, proposes that it is open to Parliament to add a new tool to its arsenal of punitive implements — denial of constitutional rights.  I find this notion problematic.  I do not doubt that Parliament may limit constitutional rights in the name of punishment, provided that it can justify the limitation.  But it is another thing to say that a particular class of people for a particular period of time will completely lose a particular constitutional right.  This is tantamount to saying that the affected class is outside the full protection of the Charter .  It is doubtful that such an unmodulated deprivation, particularly of a right as basic as the right to vote, is capable of justification under s. 1.  Could Parliament justifiably pass a law removing the right of all penitentiary prisoners to be protected from cruel and unusual punishment?  I think not.  What of freedom of expression or religion?  Why, one asks, is the right to vote different? The government offers no credible theory about why it should be allowed to deny this fundamental democratic right as a form of state punishment.

47                               The social compact requires the citizen to obey the laws created by the democratic process.  But it does not follow that failure to do so nullifies the citizen’s continued membership in the self-governing polity.  Indeed, the remedy of imprisonment for a term rather than permanent exile implies our acceptance of continued membership in the social order.  Certain rights are justifiably limited for penal reasons, including aspects of the rights to liberty, security of the person, mobility, and security against search and seizure.  But whether a right is justifiably limited cannot be determined by observing that an offender has, by his or her actions, withdrawn from the social compact.  Indeed, the right of the state to punish and the obligation of the criminal to accept punishment are tied to society’s acceptance of the criminal as a person with rights and responsibilities.  Other Charter provisions make this clear.  Thus s. 11 protects convicted offenders from unfair trials, and s. 12 from “cruel and unusual treatment or punishment”.

48                               The second flaw in the argument that s. 51(e) furthers legitimate punishment is that it does not meet the dual requirements that punishment must not be arbitrary and must serve a valid criminal law purpose. Absence of arbitrariness requires that punishment be tailored to the acts and circumstances of the individual offender: R. v. Smith, [1987] 1 S.C.R. 1045, at p. 1073.  In the immortal words of Gilbert and Sullivan, the punishment should fit the crime.  Section 51(e) qua punishment bears little relation to the offender’s particular crime.  It makes no attempt to differentiate among inmates serving sentences of two years and those serving sentences of twenty.  It is true that those serving shorter sentences will be deprived of the right to vote for a shorter time.  Yet the correlation of the denial with the crime remains weak.  It is not only the violent felon who is told he is an unworthy outcast; a person imprisoned for a non-violent or negligent act, or an Aboriginal person suffering from social displacement receives the same message.  They are not targeted, but they are caught all the same.  For them the message is doubly invidious — not that they are cast out for their apparently voluntary rejection of society’s norms, but that they are cast out arbitrarily, in ways that bear no necessary relation to their actual situation or attitude towards state authority. 

49                               Punishment must also fulfill a legitimate penal purpose:  see Smith, supra, at p. 1068.  These include deterrence, rehabilitation, retribution, and denunciation: R. v. M. (C.A.), [1996] 1 S.C.R. 500, at para. 82.  Neither the record nor common sense supports the claim that disenfranchisement deters crime or rehabilitates criminals.  On the contrary, as Mill recognized long ago, participation in the political process offers a valuable means of teaching democratic values and civic responsibility.

50                               This leaves retribution and denunciation.  Parliament may denounce unlawful conduct.  But it must do so in a way that closely reflects the moral culpability of the offender and his or her circumstances. As Lamer C.J. indicated in M. (C.A.), supra, at para. 80:

Retribution in a criminal context, by contrast [to vengeance], represents an objective, reasoned and measured determination of an appropriate punishment which properly reflects the moral culpability of the offender, having regard to the intentional risk-taking of the offender, the consequential harm caused by the offender, and the normative character of the offender’s conduct. [Emphasis in original.]

Denunciation as a symbolic expression of community values must be individually tailored in order to fulfill the legitimate penal purpose of condemning a particular offender’s conduct (see M. (C.A.), supra, at para. 81) and to send an appropriate “educative message” about the importance of law-abiding behavior. 

51                               Section 51(e) imposes blanket punishment on all penitentiary inmates regardless of the particular crimes they committed, the harm they caused, or the normative character of their conduct.  It is not individually tailored to the particular offender’s act.  It does not, in short, meet the requirements of denunciatory, retributive punishment. It follows that it is not rationally connected to the goal of imposing legitimate punishment.

52                               When the facade of rhetoric is stripped away, little is left of the government’s claim about punishment other than that criminals are people who have broken society’s norms and may therefore be denounced and punished as the government sees fit, even to the point of removing fundamental constitutional rights.  Yet, the right to punish and to denounce, however important, is constitutionally constrained.  It cannot be used to write entire rights out of the Constitution, it cannot be arbitrary, and it must serve the constitutionally recognized goals of sentencing.  On all counts, the case that s. 51(e) furthers lawful punishment objectives  fails.

53                               I conclude that the government has failed to establish a rational connection between s. 51(e)’s denial of the right to vote and the objectives of enhancing respect for the law and ensuring appropriate punishment. 

2.  Minimal Impairment

54                               If the denial of a right is not rationally connected to the government’s objectives, it makes little sense to go on to ask whether the law goes further than is necessary to achieve the objective.  I simply observe that if it were established that denying the right to vote sends an educative message that society will not tolerate serious crime, the class denied the vote — all those serving sentences of two years or more — is too broad, catching many whose crimes are relatively minor and who cannot be said to have broken their ties to society.  Similarly, if it were established that this denial somehow furthers legitimate sentencing goals, it is plain that the marker of a sentence of two years or more catches many people who, on the government’s own theory, should not be caught. 

55                               The question at this stage of the analysis is not how many citizens are affected, but whether the right is minimally impaired.  Even one person whose Charter  rights are unjustifiably limited is entitled to seek redress under the Charter .  It follows that this legislation cannot be saved by the mere fact that it is less restrictive than a blanket exclusion of all inmates from the franchise.  First, it is difficult to substantiate the proposition that a two-year term is a reasonable means of identifying those who have committed “serious”, as opposed to “minor”, offences.  If serious and minor offences are defined by the duration of incarceration, then this is a tautology. If the two-year period is meant to serve as a proxy for something else, then the government must give content to the notion of “serious” vs. “minor” offences, and it must demonstrate the correlation between this distinction and the entitlement to vote.  It is no answer to the overbreadth critique to say that the measure is saved because a limited class of people is affected: the question is why individuals in this class are singled out to have their rights restricted, and how their rights are limited.  The perceived “seriousness” of the crime is only one of many factors in determining the length of a convicted offender’s sentence and the time  served.  The only real answer the government provides to the question “why two years?” is because it affects a smaller class than would a blanket disenfranchisement.

56                               Nor is it any answer to say that the infringement will end when the imprisonment ends.  The denial of the right to vote during the period of imprisonment affects penitentiary inmates consistently, to an absolute degree, and in arbitrary ways that bear no necessary relation to their actual situation or attitude towards state authority.  Section 51(e) thus denies a prisoner’s rights in precisely the same fashion as its unconstitutional predecessor.

3.  Proportionate Effect

57                               If a connection could be shown between the denial of the right to vote and the government’s objectives, the negative effects of denying citizens the right to vote would greatly outweigh the tenuous benefits that might ensue.  

58                               Denial of the right to vote to penitentiary inmates undermines the legitimacy of government, the effectiveness of government, and the rule of law.  It curtails the personal rights of the citizen to political expression and participation in the political life of his or her country.  It countermands the message that everyone is equally worthy and entitled to respect under the law — that everybody counts: see August, supra.  It is more likely to erode respect for the rule of law than to enhance it, and more likely to undermine sentencing goals of deterrence and rehabilitation than to further them.

59                               The government’s plea of no demonstrated harm to penitentiary inmates rings hollow when what is at stake is the denial of the fundamental right of every citizen to vote.  When basic political rights are denied, proof of additional harm is not required.  But were proof needed, it is available.  Denying prisoners the right to vote imposes negative costs on prisoners and on the penal system.  It removes a route to social development and rehabilitation acknowledged since the time of Mill, and it undermines correctional law and policy directed towards rehabilitation and integration (testimony of Professor Jackson, appellants’ record at pp. 2001-2).  As the trial judge clearly perceived at p. 913, s. 51(e) “serves to further alienate prisoners from the community to which they must return, and in which their families live”.

60                               The negative effects of s. 51(e) upon prisoners have a disproportionate impact on Canada’s already disadvantaged Aboriginal population, whose overrepresentation in prisons reflects “a crisis in the Canadian criminal justice system”: R. v. Gladue, [1999] 1 S.C.R. 688, at para. 64, per Cory and Iacobucci JJ.  To the extent that the disproportionate number of Aboriginal people in penitentiaries reflects factors such as  higher rates of poverty and institutionalized alienation from mainstream society, penitentiary imprisonment may not be a fair or appropriate marker of the degree of individual culpability.  Added to this is the cost of silencing the voices of incarcerated Aboriginal people; with due respect, the fact that 1,837 Aboriginal people are disenfranchised by this law, while close to 600,000 are not directly affected, does not justify restricting the rights of those 1,837 individuals for reasons not demonstrably justified under the Charter : see Court of Appeal decision at para. 169.  Aboriginal people in prison have unique perspectives and needs.  Yet, s. 51(e) denies them a voice at the ballot box and, by proxy, in Parliament.  That these costs are confined to the term of imprisonment does not diminish their reality.  The silenced messages cannot be retrieved, and the prospect of someday participating in the political system is cold comfort to those whose rights are denied in the present.

61                               In the final analysis, even if there were merit in the  Court of Appeal’s view that the trial judge relied too heavily on the absence of concrete evidence of benefit, it is difficult to avoid the trial judge’s conclusion, at p. 916, that “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 of proof”.

62                               I conclude that s. 51(e)’s disenfranchisement of prisoners sentenced to two years or more cannot be justified under s. 1  of the Charter . I leave for another day whether some political activities, like standing for office, could be justifiably denied to prisoners under s. 1.  It may be that practical problems might serve to justify some limitations on the exercise of derivative democratic rights.  Democratic participation is not only a matter of theory but also of practice, and legislatures retain the power to limit the modalities of its exercise where this can be justified.  Suffice it to say that the wholesale disenfranchisement of all penitentiary inmates, even with a two-year minimum sentence requirement, is not demonstrably justified in our free and democratic society.

D.  The Guarantee of Equality under Section 15(1)  of the Charter 

63                               Having found that s. 51(e) unjustifiably infringes s. 3  of the Charter , it is unnecessary to consider the alternative argument that it infringes the equality guarantee of s. 15(1).

V.  Conclusion

64                               I would allow the appeal, with costs to the appellants.  Section 51(e) infringes s. 3  of the Charter , and the infringement is not justified under s. 1.  It follows that s. 51(e) is inconsistent with the Charter  and is of no force or effect by operation of s. 52  of the Constitution Act ,  1982 .  I would answer the constitutional questions as follows:

1.                Does s. 51(e) of the Canada Elections Act, R.S.C. 1985, c. E-2, infringe the right to vote in an election of members of the House of Commons, as guaranteed by s. 3  of the Canadian Charter of Rights and Freedoms ?

Yes.

2.                If the answer to Question 1 is yes, is the infringement a reasonable limit, prescribed by law, which can be demonstrably justified in a free and democratic society, pursuant to s. 1  of the Canadian Charter of Rights and Freedoms ?

No.

It is unnecessary to answer the constitutional questions regarding s. 15(1)  of the Charter .

The reasons of L’Heureux-Dubé, Gonthier, Major and Bastarache JJ. were delivered by

Gonthier J. (dissenting) —

I.  Introduction

A)  Specific Issue Before this Court

65                               Is Parliament able to temporarily suspend the exercise of the right to vote for criminals incarcerated for the commission of serious crimes for the duration of their incarceration?  This is the question raised by this appeal.  The answer will depend on whether s. 51(e) of the Canada Elections Act, R.S.C. 1985, c. E-2 (the “Act”), which prohibits “[e]very person who is imprisoned in a correctional institution serving a sentence of two years or more” from voting, is in breach of ss. 3  or 15  of the Canadian Charter of Rights and Freedoms  in a manner not justifiable under s. 1. 

66                               The trial judge was of the view that s. 51(e) of the Act did not satisfy the test mandated by s. 1  of the Charter .  I am in respectful disagreement with the reasons of my colleague, Chief Justice McLachlin, which support the disposition reached by the trial judge.  I generally agree with the reasoning of the majority of the Federal Court of Appeal below that this provision is constitutionally sound.  In my view, s. 51(e) of the Act is not an infringement of s. 15  of the Charter , and while having been conceded to be an infringement of the s. 3  Charter  right, it is capable of being justified under s. 1  of the Charter  as a reasonable limitation thereupon.

B)  The More Fundamental Issue Arising from the Context of the Case at Bar

67                               My disagreement with the reasons of the Chief Justice, however, is also at a  more fundamental level.  This case rests on philosophical, political and social considerations which are not capable of “scientific proof”.  It involves justifications for and against the limitation of the right to vote which are based upon axiomatic arguments of principle or value statements.  I am of the view that when faced with such justifications, this Court ought to turn to the text of s. 1  of the Charter  and to the basic principles which undergird both s. 1 and the relationship that provision has with the rights and freedoms protected within the Charter .  Particularly, s. 1  of the Charter  requires that this Court look to the fact that there may be different social or political philosophies upon which justifications for or against the limitations of rights may be based.  In such a context, where this Court is presented with competing social or political philosophies relating to the right to vote, it is not by merely approving or preferring one that the other is necessarily disproved or shown not to survive Charter  scrutiny.  If the social or political philosophy advanced by Parliament reasonably justifies a limitation of the right in the context of a free and democratic society, then it ought to be upheld as constitutional.  I conclude that this is so in the case at bar.

II.  Legislative Provision in Question

68                               I am of the view that by enacting s. 51(e) of the Act, Parliament has chosen to assert and enhance the importance and value of the right to vote by temporarily disenfranchising serious criminal offenders for the duration of their incarceration.  This point is worth underlining. The Chief Justice and I are in agreement that the right to vote is profoundly important, and ought not to be demeaned. Our differences lie principally in the fact that she subscribes to a philosophy whereby the temporary disenfranchising of criminals does injury to the rule of law, democracy and the right to vote, while I prefer deference to Parliament’s reasonable view that it strengthens these same features of Canadian society. 

69                               The reasons of the Chief Justice refer to the historical evolution of the franchise in Canada.  This evolution has generally involved the weeding out of discriminatory exclusions.  It is undeniable and, obviously, to be applauded, that, over time, Canada has been evolving towards the universalization of the franchise in such a manner.  The provision in question in the case at bar, however, is strikingly and qualitatively different from these past discriminatory exclusions. It is a temporary suspension from voting based exclusively on the serious criminal activity of the offender.  It is the length of the sentence, reflecting the nature of the offence and the criminal activity committed, that results in the temporary disenfranchisement during incarceration.  Thus, far from being repugnant and discriminatory, based on some irrelevant personal characteristic, such as gender, race, or religion, s. 51(e) of the Act distinguishes persons based on the perpetrating of acts that are condemned by the Criminal Code , R.S.C. 1985, c. C-46 .  Parliament has recognized this distinction as being different from other exclusions by its continued assertion that being convicted of a serious criminal offence is a ground for temporary disenfranchisement.

70                               The reasons of the Chief Justice at para. 33 cite with approval a passage from Arbour J.A. ’s reasons in the first Sauvé case which suggests that:

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. 

(Sauvé v. Canada (Attorney General) (1992), 7 O.R. (3d) 481 (C.A.), at p. 487)

While there is little logical correlation between maintaining a “decent and responsible citizenry” and any of the past discriminatory exclusions (such as land-ownership, religion, gender, ethnic background), there clearly is such a logical connection in the case of distinguishing persons who have committed serious criminal offences. “Responsible citizenship” does not relate to what gender, race, or religion a person belongs to, but is logically related to whether or not a person engages in serious criminal activity.

71                               A further dimension of this qualitative difference is that serious criminal offenders are excluded from the vote for the reason that they are the subjects of punishment.  The disenfranchisement only lasts as long as the period of incarceration.  Thus, disenfranchisement, as a dimension of punishment, is attached to and mirrors the fact of incarceration.  This fact makes the Canadian experience significantly different from the situation in some American states which disenfranchise ex-offenders for life, a situation addressed by many American academics: see, for example, L. H. Tribe, “The Disenfranchisement of Ex-Felons: Citizenship, Criminality, and ‘The Purity of the Ballot Box’” (1989), 102 Harv. L. Rev. 1300.

72                               It is important to look at prisoner disenfranchisement from the perspective of each serious criminal offender rather than perceive it as a form of targeted group treatment.  Disenfranchised prisoners can be characterized loosely as a group, but what is important to realize is that each of these prisoners has been convicted of a serious criminal offence and is therefore serving a personalized sentence which is proportionate to the act or acts committed.  Punishment is guided by the goals of denunciation, deterrence, rehabilitation and retribution and is intended to be morally educative for incarcerated serious criminal offenders.  Each prisoner’s sentence is a temporary measure aimed at meeting these goals, while also being aimed at the long-term objective of reintegration into the community.

73                               The reasons of the Chief Justice express the view that the temporary disenfranchisement of serious criminal offenders necessarily undermines their inherent “worth” or “dignity.”  I disagree.  In fact, it could be said that the notion of punishment is predicated on the dignity of the individual: it recognizes serious criminals as rational, autonomous individuals who have made choices.  When these citizens exercise their freedom in a criminal manner, society imposes a concomitant responsibility for that choice.  As Professor J. Hampton, one of the Crown’s experts, writes in an article cited by Linden J.A. below, “Punishment, Feminism, and Political Identity:  A Case Study in the Expressive Meaning of the Law” (1998), 11 Can. J. L. & Jur. 23, at p. 43:

By telling people “you can have your right to vote suspended if, through your actions, you show contempt for the values that make our society possible”, this law links the exercise of freedom with responsibility for its effects.  Indeed, not to construct a punishment that sends this message is . . . to indirectly undermine the values of a democratic society.  [Underlining added; italics in original.]

74                               If there is any negative connotation associated with this temporary disenfranchisement, it arises from the fact that a criminal act was perpetrated, an act for which the criminal offender is consequently being punished.  This is not stereotyping.  Criminal acts are rightly condemned by society.  Serious criminals being punished and temporarily disenfranchised are not in any way of less “worth” or “dignity” because social condemnation is of the criminal acts and its purpose is not to diminish the individual prisoner as a person.

75                               The argument that the temporary disenfranchisement of serious criminal offenders undermines the inherent “worth” or “dignity” of prisoners presents a potentially problematic line of reasoning.  Is it possible to “punish” serious criminals without undermining their “worth”?  It  must be so. This is inherently recognized in the Charter  itself insofar that s. 12  only renders unconstitutional punishment that is “cruel and unusual”.  The Criminal Code  and its provisions are declaratory of values, values on which Canadian society rests: see R. v. Keegstra, [1990] 3 S.C.R. 697, at pp. 769 and 787.  Protecting and enhancing these values through the imposition of punishment for criminal activity is not an affront to dignity.  On the contrary, the temporary disenfranchisement of serious criminal offenders reiterates society’s commitment to the basic moral values which underpin the Criminal Code ; in this way it is morally educative for both prisoners and society as a whole.

76                               The punishment of serious criminal offenders is also aimed at protecting society and the “dignity” and “worth” of those members of society who have been or may become the victims of crime.  Punishment is intended to act as a general deterrent to potential criminals and as a specific deterrent vis-à-vis incarcerated persons.  Charter  analysis is meant to consider the Charter  rights of other members of society: see R. v. Sharpe, [2001] 1 S.C.R. 45, 2001 SCC 2, at para. 187; Keegstra, supra, at p. 756.  Serious criminal activity is clearly often an affront to numerous Charter  values.

III.  The Right to Vote in Section 3  of the Charter 

77                               The current status of the vote in Canada is that all citizens are empowered to vote in federal elections, save incarcerated offenders who are disenfranchised by s. 51(e) of the Act (now continued in substantially the same form at s. 4(c) of the Act  (S.C. 2000, c. 9)), the Chief and Assistant Chief Electoral Officers, and persons under the age of 18 as of polling day.  The most recent changes to the scope of the franchise were made by Parliament in 1993, when the disqualifications of federally appointed judges and persons suffering from “mental disease” were removed: S.C. 1993, c. 19,  s. 23 amending s. 51 of the Act, and s. 24 which added s. 51.1.  These modifications included s. 51(e), which disqualified incarcerated persons serving sentences of two years or more.  It was already enacted at the time this Court heard the first Sauvé case, [1993] 2 S.C.R. 438 (the “first Sauvé case”), but since it was inapplicable to the case under appeal, the Court did not comment on it.

78                               The respondents conceded that s. 3  of the Charter  has been infringed by s. 51(e) of the Act.  I would like to sound a cautionary note regarding the appropriateness of concessions of infringement.  The specific problem with such a concession is that it may deprive the courts of the benefit of the fruitful argument which most often occurs at that initial phase of the analysis, in defining the scope of the right, particularly with regard to historical and philosophical context.  The development of contextual factors examined with regard to the scope of the right is of great importance since they clearly “animate” the later stages of the test elaborated in R. v. Oakes, [1986] 1 S.C.R. 103: see McLachlin J. (as she then was) in Reference re Provincial Electoral Boundaries (Sask.), [1991] 2 S.C.R. 158, at p. 182.  The following comment, which I shared with L’Heureux-Dubé and Bastarache JJ. in Sharpe, supra, at para. 151, is equally apposite here: “it is unfortunate that the Crown conceded that the right to free expression was violated in this appeal in all respects, thereby depriving the Court of the opportunity to fully explore the content and scope of s. 2( b )  as it applies in this case”.

IV.  Section 1  of the Charter 

A)  Oakes: Flexibility and Context

79                               To decide whether the limit upon the s. 3  Charter  right found in s. 51(e) of the Act is justified, one must determine whether it constitutes a “reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society”.  Pursuant to the test formulated in Oakes, supra, this analysis proceeds in two stages.  At the preliminary phase, it must be determined whether the objective behind the limit is of sufficient importance to justify overriding a Charter  right. At the second stage, we must consider whether the legislative measure chosen is rationally connected to the legislative objective, whether the measure minimally impairs the Charter  right which has been infringed, and finally whether the effects of the measure are proportional to the significance of the objective and whether the salutary and deleterious effects of the measure are proportional.

80                               When engaging in Charter  analysis, context and flexibility are highly relevant.  The need for flexibility in the application of the Oakes test was noted by La Forest J. in United States of America v. Cotroni, [1989] 1 S.C.R. 1469, at pp. 1489-90:

In the performance of the balancing task under s. 1, it seems to me, a mechanistic approach must be avoided.  While the rights guaranteed by the Charter  must be given priority in the equation, the underlying values must be sensitively weighed in a particular context against other values of a free and democratic society sought to be promoted by the legislature.

Dickson C.J. adopted the above passage in his majority reasons in Keegstra, supra, and he went on to state at pp. 737-38:

. . . I hope it is clear that a rigid or formalistic approach to the application of s. 1 must be avoided.  The ability to use s. 1 as a gauge which is sensitive to the values and circumstances particular to an appeal has been identified as vital in past cases . . . .  The sentiments of La Forest J. correctly suggest that the application of the Oakes approach will vary depending on the circumstances of the case, including the nature of the interests at stake.

81                               Factual, social, historical, and political context provides a backdrop which is essential to develop in order to properly analyse what is at stake in the case of an alleged infringement of a right.  As Bastarache J. noted in Thomson Newspapers Co.  v. Canada (Attorney General), [1998] 1 S.C.R. 877, at para. 87:

The analysis under s. 1  of the Charter  must be undertaken with a close attention to context.  This is inevitable as the test devised in R. v. Oakes, [1986] 1 S.C.R. 103, requires a court to establish the objective of the impugned provision, which can only be accomplished by canvassing the nature of the social problem which it addresses.  Similarly, the proportionality of the means used to fulfil the pressing and substantial objective can only be evaluated through a close attention to detail and factual setting.  In essence, context is the indispensable handmaiden to the proper characterization of the objective of the impugned provision, to determining whether that objective is justified, and to weighing whether the means used are sufficiently closely related to the valid objective so as to justify an infringement of a Charter  right.  [Emphasis added.]

See also Sharpe, supra, at paras. 132 and 156.

82                               L’Heureux-Dubé and Bastarache JJ. and myself in Sharpe, supra, at para. 153, noted that it is by virtue of attention to context that courts come to consider other communal values which must be addressed in s. 1 analysis:

While the guidelines set out in Oakes provide a useful analytical framework for the practical application of s. 1, it is important not to lose sight of the underlying purpose of that section, namely to balance individual rights and our communal values.  Where courts are asked to consider whether a violation is justified under s. 1, they must be sensitive to the competing rights and values that exist in our democracy.

These other communal values have a legitimate role to play in Charter  analysis. As stated by Dickson C.J. in Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038, at p. 1056: “There are many diverse values that deserve protection in a free and democratic society such as that of Canada, only some of which are expressly provided for in the Charter .”

83                               In the case at bar, given that the context involves evaluating choices regarding social or political philosophies and relates to shaping, giving expression, and giving practical application to communal values, it is of the utmost significance.

B)  Section 3 is Subject to Section 1

84                               The flexible contextual approach to s. 1  of the Charter  and the Oakes test necessitates that this Court keep in mind first principles.  The right to vote for all citizens is clearly encapsulated in s. 3  of the Charter  and, by the terms of s. 1, it is subject to “such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society”.  The right, therefore, is not absolute, but is qualified — as are all rights under the Charter  — by s. 1.  A violation of the Charter  only manifests after the s. 1 inquiry has been entertained, since s. 1 permits and envisions that the rights and freedoms enshrined in the Charter  are capable of such limitation.

85                               An examination of the development of the language of s. 3 when the wording of the Charter  was being reviewed by the Special Joint Committee on the Constitution evidences Parliament’s intent that s. 3 must be read as subject to s. 1 limitations.  Professor P. W. Hogg, in Constitutional Law of Canada (loose-leaf ed.), offers the following commentary regarding an earlier version of the language of s. 3, at p. 42-2, fn. 12:

Section 3  of the Charter  was in the October 1980 version in the following terms:

3. Every citizen of Canada has, without unreasonable distinction or limitation, 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.

The April 1981 version contained the final version of s. 3. In the final version the words “without unreasonable distinction or limitation” are deleted. The reason for the deletion was, no doubt, that the words were redundant having regard to s. 1. [Emphasis added.]

The Minutes of Proceedings and Evidence of the Special Joint Committee on the Constitution of Canada, Issue No. 43, January 22, 1981, at pp. 43:79 - 43:90, support this statement:  that Committee decided to remove the built-in limitation within s. 3 simply because s. 1 was intended to apply to s. 3 and allow demonstrably justifiable limitations to survive constitutional scrutiny.

86                               This general approach to s. 3 does not mean that justificatory concerns play a role in defining the content of s. 3, or that the right to vote necessarily has inherent limitations within it.  This proposition was rejected by the majority of this Court in Harvey v. New Brunswick (Attorney General), [1996] 2 S.C.R. 876, at paras. 29-32.  It does mean that the specific context needs to be brought into the s. 1  Charter  test for the balancing of interests, since s. 3, like all Charter  rights, is not absolute.  This is similar to the approach advocated in R. v. Mills, [1999] 3 S.C.R. 668, at para. 61, where McLachlin J. and Iacobucci J. stressed:

. . . the importance of interpreting rights in a contextual manner — not because they are of intermittent importance but because they often inform, and are informed by, other similarly deserving rights or values at play in particular circumstances.

87                               To graphically illustrate the above as it might be applied to prisoners, one may ask the question: could a prisoner argue successfully that, by reason of his incarceration, there was a violation of s. 6(1)  of the Charter , which guarantees that “[e]very citizen of Canada has the right to enter, remain in and leave Canada”?  It would border on the ludicrous to suggest that a limitation on this right was not possible under s. 1  of the Charter . 

88                               In the same vein, what of the s. 3  Charter  guarantee that every citizen has a right to be qualified for membership in the House of Commons?  Must a prisoner have the right to stand as a candidate?  Were an incarcerated offender to be elected, would he or she have a right to be released from prison to take up that representative role?

89                               To construe s. 3 as an absolute right presents other problems.  For example, currently only persons of the age of 18 or older are permitted to vote.  How, then, can there be a justification for denying the vote to a politically mature 16 or 17 year-old?  To justify this line, would Parliament have to resort to the argument that persons below the age of majority are not, in some sense, citizens in the fullest sense?  The answer must simply lie in the fact that s. 1 allows Parliament to make such choices as long as they are rational and reasonable limitations which are justified in a free and democratic society.

C)  Justification Under Section 1 Where the Context Involves Competing Social or

     Political Philosophies

90                               Generally, the result of a concession of a breach of a right is that the burden for justifying the infringement falls to the Crown.  With regard to the standard of proof for s. 1 analysis, the civil standard of the balance of probabilities applies, and it is important to stress that this standard does not require scientific proof: Libman v. Quebec (Attorney General), [1997] 3 S.C.R. 569, at para. 39.  The application of common sense to what is known will suffice, even if this falls short of a scientific standard: McLachlin J. at para. 137 of RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199.  Justification can be based upon reason, logic, common sense, and knowledge of human nature: see R. v. Butler, [1992] 1 S.C.R. 452, at p. 524, referring to Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927, and Keegstra, supra.  In the case at bar, there is very little quantitative or empirical evidence either way.  In such cases, the task of justification relates to the analysis of human motivation, the determination of values, and the understanding of underlying social or political philosophies — it truly is justification rather than measurement.

91                               Justifying, therefore, is not a matter of one value clearly prevailing over the other, but is rather a matter of developing the significance of the values being dealt with and asking whether Parliament, in its attempt to reconcile competing interests,  has achieved a rational and reasonable balance.  Proportionality, in the context of Charter  analysis, does not mean a perfect solution, as any balance arising from competing interests will involve preferring one value over the other to some extent.

92                               As emerges from the submissions before this Court, there seem generally to be two options available for dealing with the issue at hand.  The first, that chosen by the Chief Justice, is to prefer an inclusive approach to democratic participation for serious criminal offenders incarcerated for two years or more.  This view locates democratic participation as a central dimension of rehabilitation, insofar as the incarcerated offenders remain citizens with the fullest exercise of their democratic rights.  By the same token, the unrestricted franchise enhances democratic legitimacy of government, and confirms or enhances the citizenship or standing of prisoners in society.  To do otherwise, it is suggested, undermines the “dignity” or “worth” of prisoners.  The alternative view, adopted by Parliament, considers that the temporary suspension of the prisoner’s right to vote, in fact, enhances the general purposes of the criminal sanction, including rehabilitation. It does so by underlining the importance of civic responsibility and the rule of law.  This approach sees the temporary removal of the vote as a deterrent to offending or re-offending and the return of the vote as an inducement to reject further criminal conduct.  In withdrawing for a time one expression of political participation concurrently with personal freedom, the significance of both are enhanced.  Rather than undermine the dignity or worth of prisoners, the removal of their vote takes seriously the notion that they are free actors and attaches consequences to actions that violate certain core values as expressed in the Criminal Code .

93                               Both of these approaches, however, entail accepting logically prior political or social philosophies about the nature and content of the right to vote.  The former approach, that accepted by the reasons of the Chief Justice, entails accepting a philosophy that preventing criminals from voting does damage to both society and the individual, and undermines prisoners’ inherent worth and dignity.  The latter approach also entails accepting a philosophy, that not permitting serious incarcerated criminals to vote is a social rejection of serious crime which reflects a moral line which safeguards the social contract and the rule of law and bolsters the importance of the nexus between individuals and the community.  Both of these social or political philosophies, however, are aimed at the same goal:  that of supporting the fundamental importance of the right to vote itself.  Further, both of these social or political philosophies are supported by the practices of the various Canadian provinces, the practices of other liberal democracies, and academic writings.  Finally, neither position can be proven empirically — rather, the selection of one over the other is largely a matter of philosophical preference.  What is key to my approach is that the acceptance of one or the other of these social or political philosophies dictates much of the constitutional analysis which ensues, since the reasonableness of any limitation upon the right to vote and the appropriateness of particular penal theories and their relation to the right to vote will logically be related to whether or not the justification for that limitation is based upon an “acceptable” social or political philosophy.

94                               The reasons of the Chief Justice hold, at para. 18, that the challenge of the government is to present a justification that is “convincing, in the sense that it is sufficient to satisfy the reasonable person looking at all the evidence and relevant considerations, that the state is justified in infringing the right at stake to the degree it has”.  I agree with this test, subject only to a recognition that as the context of the case at bar involves evaluating competing social or political philosophies, the analysis runs the risk of lapsing into the realm of ipse dixit. In the realm of competing social or political philosophies, reasonableness is the predominant s. 1 justification consideration.

95                               The reasons of the Chief Justice apply something seemingly more onerous than the “justification” standard referred to just above.  She describes the right to vote as a “core democratic right” and suggests that its exemption from the s. 33 override somehow raises the bar for the government in attempting to justify its restriction  (paras. 13 and 14).  This altering of the justification standard is problematic in that it seems to be based upon the view that there is only one plausible social or political philosophy upon which to ground a justification for or against the limitation of the right.  This approach, however, is incorrect on a basic reading of s. 1  of the Charter , which clearly does not constrain Parliament or authorize this Court to prioritize one reasonable social or political philosophy over reasonable others, but only empowers this Court to strike down those limitations which are not reasonable and which cannot be justified in a free and democratic society.

96                               The analysis cannot be skirted by qualifying the right to vote as a core democratic right.  It does not follow from the fact that Parliament is denied the authority to remove or qualify the right to vote in its sole discretion under s. 33 that limitations on that right may not be justified under s. 1, or that a more onerous s. 1 analysis must necessarily apply. Constitutional writers and commentators point out that s. 33 was a political compromise, meant to bring together provinces opposed to the entrenchment of constitutional rights, with those in favour:  P. Macklem et al., Canadian Constitutional Law (2nd ed. 1997), at pp. 597 and 646.  Indeed Macklem et al. write at p. 597:  “Added to the Charter  at the last moment, this controversial provision captured the final political compromise among the provinces and federal government that facilitated the adoption of the Charter ”.  There is little evidence of the intention behind excluding democratic rights (along with mobility rights, language rights, and enforcement provisions) from the ambit of s. 33 , nor has this Court ever seriously considered the significance of such exclusion.  The Chief Justice’s conclusion at para. 11 that “[t]he framers of the Charter  signaled the special importance of this right . . . by exempting it from legislative override . . .” requires examination before it can be used as support for nearly insulating the right to vote from s. 1 limitations.  In fact, s. 33  and s. 1 are clearly different in their purpose, and the Charter  clearly distinguishes their application to the right to vote.  It does not behoove the Court to read s. 33  into s. 3 by finding in s. 3, when divorced from s. 1, the statement of a political philosophy which preempts another political philosophy which is reasonable and justified under the latter section.  The Charter was not intended to monopolize the ideological space.

97                               There is a flaw in an analysis which suggests that because one social or political philosophy can be justified, it necessarily means that another social or political philosophy is not justified: in other words, where two social or political philosophies exist, it is not by approving one that you disprove the other.  Differences in social or political philosophy, which result in different justifications for limitations upon rights, are perhaps inevitable in a pluralist society.  That having been said, it is only those limitations which are not reasonable or demonstrably justified in a free and democratic society which are unconstitutional.  Therefore, the most significant analysis in this case is the examination of the social or political philosophy underpinning the justification advanced by the Crown.  This is because it will indicate whether the limitation of the right to vote is reasonable and is based upon a justification which is capable of being demonstrated in a free and democratic society.  If the choice made by Parliament is such, then it ought to be respected.  The range of choices made by different legislatures in different jurisdictions, which I will review below, supports the view that there are many resolutions to the particular issue at bar which are reasonable; it demonstrates that there are many possible rational balances.

98                               The role of this Court, when faced with competing social or political philosophies and justifications dependent on them, is therefore to define the parameters within which the acceptable reconciliation of competing values lies.  The decision before this Court is therefore not whether or not Parliament has made a proper policy decision, but whether or not the policy position chosen is an acceptable choice amongst those permitted under the Charter .  This was the view advanced by Linden J.A. for the Court of Appeal ([2000] 2 F.C. 117, at para. 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. [Emphasis in original.]

This Court has often affirmed this view that courts must be cautious not to unduly interfere in decisions which involve the balancing of conflicting policy considerations: see Reference re Public Service Employee Relations Act (Alta.), [1987] 1 S.C.R. 313, per Le Dain J., at p. 392; Canada v. Schmidt, [1987] 1 S.C.R. 500, per La Forest J., at pp. 522-23; Canada (Auditor General) v. Canada (Minister of Energy, Mines and Resources), [1989] 2 S.C.R. 49, per Dickson C.J., at pp. 90-92.

D)  Symbolic Arguments and Evidentiary Problems

99                               A subject that is related to and follows from the above discussion concerning the evaluation of competing social or political philosophies is the role of symbolic arguments in Charter  adjudication.  In the context of the Charter  analysis, it is important  not to downplay the importance of symbolic or abstract arguments. Symbolic or abstract arguments cannot be dismissed outright by virtue of their symbolism:  many of the great principles, the values upon which society rests, could be said to be symbolic.  In fact, one of the more important dimensions of s. 3  of the Charter  is clearly its symbolism:  the affirmation of political equality reflected in all citizens being guaranteed the right to vote, subject only to reasonable limits prescribed by law that can be demonstrably justified in a free and democratic society.  The case at bar concerns debates about symbolism, as the arguments involved relate to abstract concepts such as democracy, rights, punishment, the rule of law and civic responsibility. To choose a narrow reading of rights over the objectives advanced by Parliament is to choose one set of symbols over another.

100                           In her reasons, the Chief Justice claims at para. 16 that Parliament is relying on “lofty objectives”, and suggests at para. 23 that the presence of “symbolic and abstract” objectives is problematic.  However, the reasons of the Chief Justice have the very same objective — to protect the value of the right to vote and the rule of law — and rely on equally vague concepts.  Breaking down the meaning and value of the right to vote, one is unavoidably led to abstract and symbolic concepts such as the rule of law, the legitimacy of law and government, and the meaning of democracy.  The Chief Justice discusses these concepts at length, along with theories of individual motivation.  For instance, relying on the philosopher J. S. Mill, she suggests at para. 38 that “[t]o deny prisoners the right to vote is to lose an important means of teaching them democratic values and social responsibility”.  This type of statement is as symbolic, abstract and philosophical as the government’s claim that denying serious incarcerated criminals the right to vote will strengthen democratic values and social responsibility.

101                           Most of the evidence in this case is that of expert opinions on the matters of political theory, moral philosophy, philosophy of law, criminology, correctional policy, and penal theory.  I would suggest distinguishing between two kinds of expert evidence in this case.  First, there is very limited social scientific evidence, e.g. in the field of criminology, that seeks to establish the practical or empirical consequences of maintaining or lifting the ban on prisoner voting.  Second, there is copious expert testimony in the nature of legal and political philosophy.  I do not think that the Court need necessarily defer to this second type of expertise, or take into account the “skill” and “reputation” of the experts in weighing this evidence (as the trial judge purported to do at [1996] 1 F.C. 857, at pp. 865-66).  First, most if not all of the philosophers or theorists on which these experts rely never in fact even addressed the specific issue of prisoner enfranchisement or disenfranchisement.  Second, legal theory expert testimony in this context essentially purports to justify axiomatic principles.  Therefore, these arguments are either persuasive or not.  In this context, it is appropriate for courts to look not only to such theoretical arguments but also beyond, to factors such as the extent of public debate on an issue, the practices of other liberal democracies and, most especially, to the reasoned view of our democratically elected Parliament.

102                           The evidence in this case, offered by both the appellants and the Crown, is abstract and symbolic and does not lend itself to being easily demonstrated.  For example, it was submitted before this Court that the Crown ought to have to demonstrate actual benefits or actual effectiveness of the provision which Parliament has chosen.  On the facts, this is a nearly impossible task.  The same demand, however, if made with regard to the effectiveness of the Criminal Code  in general or in its specific provisions, would raise similar challenges:  can it be shown that the Criminal Code  is generally effective?  It is as if to say that because there are still criminals, we ought to do away with the Criminal Code , because the existence of crime itself points towards an effectiveness problem.  Symbolic or abstract arguments must be examined seriously for what they are, because that is effectively all that is before this Court.  Further, one must not deny that the choices between and the interpretation of these symbolic or abstract arguments are clearly connected to significant concrete effects.

103                           A key justification before this Court, to be analysed in depth below, is that serious crime reflects contempt for the rule of law and a rejection of the basis for the operation of a free and democratic society.  This, if it is symbolic or abstract, reflects a core value of the community, a value that is reflected throughout the Criminal Code  and in the provision before us today.  As will be argued below, this value is based on a reasonable social or political philosophy.  Temporarily removing the vote from serious criminal offenders while they are incarcerated is both symbolic and concrete in effect.  Returning it on being released from prison is the same.

E) “Dialogue” and Deference

104                           Linden J.A., in the Federal Court of Appeal below, stressed the importance of deference to Parliament. In para. 56 of his reasons, he stated:

This case is another episode in the continuing dialogue between courts and legislatures on the issue of prisoner voting.  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 .  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. [Footnotes omitted.]

This Court has stressed the importance of “dialogue” in Vriend v. Alberta, [1998] 1 S.C.R. 493, at paras. 138-39, and in Mills, supra, at paras. 20, 57 and 125.  (See also P. W. Hogg and A. A. Bushell, “The Charter Dialogue Between Courts and Legislatures” (1997), 35 Osgoode Hall L.J. 75.)  I am of the view that since this case is about evaluating choices regarding social or political philosophies and about shaping, giving expression, and giving practical application to values, especially values that may lie outside the Charter  but are of fundamental importance to Canadians, “dialogue” is of particular importance.  In my view, especially in the context of the case at bar, the heart of the dialogue metaphor is that neither the courts nor Parliament hold a monopoly on the determination of values.  Importantly, the dialogue metaphor does not signal a lowering of the s. 1 justification standard. It simply suggests that when, after a full and rigorous s. 1 analysis, Parliament has satisfied the court that it has established a reasonable limit to a right that is demonstrably justified in a free and democratic society, the dialogue ends; the court lets Parliament have the last word and does not substitute Parliament’s reasonable choices with its own.

105                           Linden J.A. stressed the need for deference to Parliament’s having chosen to draw a particular line regarding which criminal offences are serious enough to warrant the loss of the vote.  I suggest that regardless of the relationship between the timing of the Royal Assent to the modifications to s. 51(e) of the Act and the decision of this Court in the first Sauvé case, there has been, generally, “dialogue” undertaken between the courts and Parliament.  “Dialogue” has existed insofar as Parliament had been addressing, since well before the decision of the Court of Appeal below, an evaluation of the nature of the right to vote, and specifically, the issue of prisoner disenfranchisement.  This evaluation was obviously undertaken with the many cases concerning prisoner disenfranchisement that had occurred up to that point in mind, including:  Re Jolivet and The Queen (1983), 1 D.L.R. (4th) 604 (B.C.S.C.); Gould v. Canada (Attorney General), [1984] 2 S.C.R. 124, aff’g [1984] 1 F.C. 1133 (C.A.), rev’g [1984] 1 F.C. 1119 (T.D.); Lévesque v. Canada (Attorney General), [1986] 2 F.C. 287 (T.D.); Badger v. Attorney-General of Manitoba (1986), 30 D.L.R. (4th) 108 (Man. Q.B.), aff’d (1986), 32 D.L.R. (4th) 310 (Man. C.A.); Badger v. Canada (Attorney General) (1988), 55 Man. R. (2d) 211 (Q.B.), rev’d (1988), 55 D.L.R. (4th) 177 (Man. C.A.), leave to appeal refused, [1989] 1 S.C.R. v; Sauvé v. Canada (Attorney General) (1988), 66 O.R. (2d) 234 (H.C.), rev’d (1992), 7 O.R. (3d) 481 (C.A.); Belczowski v. Canada, [1991] 3 F.C. 151 (T.D.), aff’d [1992] 2 F.C. 440 (C.A.).  What is also particularly relevant is that s. 51(e) of the Act received Royal Assent on May 6, 1993, well after both the Ontario Court of Appeal handed down its decision in the first Sauvé case (March 25, 1992) and the Federal Court of Appeal handed down its decision in Belczowski (February 17, 1992).

106                           I note as well that a Royal Commission on Electoral Reform and Party Financing, the “Lortie Commission”, was established in November 1989.  The Final Report of that Commission was submitted to Cabinet in November 1991.  That report canvassed specifically the issue of the disqualification of certain groups of voters, including prison inmates.

107                           To repeat: deference does not mean that this Court must not rigorously examine the justifications presented by the Crown for s. 51(e) of the Act.  As Iacobucci J. noted in his partial dissent in Little Sisters Book and Art Emporium v. Canada (Minister of Justice), [2000] 2 S.C.R. 1120, 2000 SCC 69, at para. 221: “While deference is appropriate, our Court cannot abdicate its duty to demand that the government justify legislation limiting Charter  rights.”  This point was also discussed by McLachlin J. in RJR-MacDonald, supra, at para. 136, explaining that there is a fine line between the appropriateness of deference and the role of this Court to safeguard Charter  rights:

. . . care must be taken not to extend the notion of deference too far.  Deference must not be carried to the point of relieving the government of the burden which the Charter  places upon it of demonstrating that the limits it has imposed on guaranteed rights are reasonable and justifiable.  Parliament has its role:  to choose the appropriate response to social problems within the limiting framework of the Constitution.  But the courts also have a role:  to determine, objectively and impartially, whether Parliament’s choice falls within the limiting framework of the Constitution.  The courts are no more permitted to abdicate their responsibility than is Parliament.  To carry judicial deference to the point of accepting Parliament’s view simply on the basis that the problem is serious and the solution difficult, would be to diminish the role of the courts in the constitutional process and to weaken the structure of rights upon which our constitution and our nation is founded.

108                           Most basically, what the commentary regarding “dialogue” and deference goes to in this case is keeping in mind the importance of what was discussed above regarding the existence of competing social or political philosophies when embarking upon flexible and contextual s. 1  Charter  analysis.  In the specific factual context of the case at bar, I think that this challenge was well summarized by Lord Justice Kennedy in a recent case from the United Kingdom regarding prisoner disenfranchisement in the context of the U.K.’s incorporation in the Human Rights Act 1998 (U.K.), 1998, c. 42, of the European Convention for the Protection of Human Rights and Fundamental Freedoms (also known as the European Convention on Human Rights (“ECHR”)), 213 U.N.T.S. 221.  In Pearson v. Secretary of State for the Home Department, [2001] E.W.J. No. 1566 (QL) (Div. Ct.), Lord Justice Kennedy  stated, at para. 23:

As Parliament has the responsibility for deciding what shall be the consequences of conviction by laying down the powers and duties of a sentencing tribunal or other body it necessarily follows that lines have to be drawn, and that on subsequent examination a case can be made in favour of the line being drawn somewhere else, but in deference to the legislature courts should not easily be persuaded to condemn what has been done, especially where it has been done in primary legislation after careful evaluation and against a background of increasing public concern about crime.

F)  A Rational and Reasonable Social or Political Philosophy Underpins the

     Crown’s Justification for the Limitation of the Section 3 Right

109                           What social or political philosophy has motivated Parliament to insist on the temporary disenfranchisement of prisoners?  Is it reasonable and rational?  I suggest that, in enacting s. 51(e) of the Act and in providing a justification of that provision before the courts, Parliament has indicated that it has drawn a line.  This line reflects a moral statement about serious crime, and about its significance to and within the community.  The core of this moral statement is the denunciation of serious crime, serious antisocial acts.  Parliament has indicated that criminal conduct of such severity that it warrants imprisonment for a sentence of two years or more also carries with it the disenfranchisement of the offender for the duration of his or her incarceration. Most importantly, as I will develop below, this basis for the Crown’s justification is both rational and reasonable.

110                           The fact that the line drawn is related to sentences which flow from the commission of crimes under the Criminal Code  is of great relevance.  As noted above, this Court has held that the Criminal Code  and its provisions are declaratory of values, values on which society rests: see Keegstra, supra, at pp. 769 and 787.  Therefore, it is perfectly appropriate to look to these underlying values and their explanation to assist in the seeking of a reconciliation of the competing interests at hand.  With regard to importing such values into Charter  analysis, I refer not only to my discussion above regarding reading s. 3  of the Charter  with s. 1, but also to the oft-cited statement by Dickson C.J. for the majority in Slaight Communications, supra, at p. 1056, that  “[t]he underlying values of a free and democratic society both guarantee the rights in the Charter  and, in appropriate circumstances, justify limitations upon those rights.”

111                           In my concurring opinion in Butler, supra, I discussed the legitimate role of the state to act on the basis of morality.  I stated, at p. 522, that “I cannot conceive that the State could not legitimately act on the basis of morality.  Since its earliest Charter  pronouncements, this Court has acknowledged this possibility”.  I continued at pp. 523-24:

In a pluralistic society like ours, many different conceptions of the good are held by various segments of the population.  The guarantees of s. 2  of the Charter  protect this pluralistic diversity.  However, if the holders of these different conceptions agree that some conduct is not good, then the respect for pluralism that underlies s. 2  of the Charter  becomes less insurmountable an objection to State action . . . . In this sense a wide consensus among holders of different conceptions of the good is necessary before the State can intervene in the name of morality.  This is also comprised in the phrase “pressing and substantial”.

112                           This view of the role of morality in law has been developed by Professor J. Raz, who states in his book The Morality of Freedom (1986), at p. 133, that “. . . it is the goal of all political action to enable individuals to pursue valid conceptions of the good and to discourage evil or empty ones”.  I agree with this view.  In Making Men Moral (1993), at p. 170, Professor R. P. George describes Professor Raz’s view as the following:

. . . political theory cannot prescind from questions of individual morality — it cannot simply leave individual morality to the individual.  The principles of political morality are tightly connected to the principles that establish the moral rectitude or culpability of individual action.  He [Professor Raz] does not conclude that the state is warranted in enforcing every moral norm; but he does argue that the state cannot adopt a position of neutrality with respect to these norms.

113                           In my view, the real challenge is not justifying state activity on the basis of morality in the abstract, but determining which specific moral claims are sufficient to warrant consideration in determining the extent of Charter  rights.  As I quoted in Butler, supra, at p. 523, Professor R. Dworkin’s Taking Rights Seriously (1977) notes, at p. 255, that:

The claim that a moral consensus exists is not itself based on a poll.  It is based on an appeal to the legislator’s sense of how his community reacts to some disfavored practice.  But this same sense includes an awareness of the grounds on which that reaction is generally supported.  If there has been a public debate involving the editorial columns, speeches of his colleagues, the testimony of interested groups, and his own correspondence, these will sharpen his awareness of what arguments and positions are in the field.  He must sift these arguments and positions, trying to determine which are prejudices or rationalizations, which pre-suppose general principles or theories vast parts of the population could not be supposed to accept, and so on.

The issue is therefore identifying what amounts to a fundamental enough conception of morality. In Butler, supra, at p. 523, I developed reasoning to assist in such identification.  The first inquiry to be satisfied is that the moral claim is grounded, meaning that it “. . . must involve concrete problems such as life, harm, well-being . . . and not merely differences of opinion or of taste.  Parliament cannot restrict Charter  rights simply on the basis of dislike; this is what is meant by the expression ‘substantial and pressing’ concern”.  The second inquiry is that “. . . a consensus must exist among the population on these claims.  They must attract the support of more than a simple majority of people”.

114                           In the case at bar, the provision in question denounces serious crime, with a view to enhancing the general purposes of the criminal sanction and to promoting civic responsibility and the rule of law.  Surely such objectives qualify as reflecting a “fundamental conception of morality”.  Attention to what is “moral” is concerned “. . . with the distinction between right and wrong” (Concise Oxford Dictionary (9th ed. 1995)).  The provision in question, like the Criminal Code  in its entirety, by virtue of the focus upon criminal activity, is addressed at specifically this.

115                           The denunciation of crime and its effects on society is often explained by reference to the notion of the social contract.  The social contract is the theoretical basis upon which the exercise of rights and participation in the democratic process rests.  In my view, the social contract necessarily relies upon the acceptance of the rule of law and civic responsibility and on society’s need to promote the same.  The preamble to the Charter  establishes that “. . . Canada is founded upon principles that recognize the supremacy of God and the rule of law. . .”. In Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721, at p. 750, this Court cited with approval a passage from The Authority of Law (1979) by Professor Raz, wherein he states that  “‘The rule of law’ means literally what it says . . . . It has two aspects:  (1) that people should be ruled by the law and obey it, and (2) that the law should be such that people will be able to be guided by it.”  The important point arising from that passage is the corollary that promoting law-abiding behaviour can be thought to be a dimension of the rule of law as well.  Further, the rule of law, as was said in the Reference re Secession of Quebec, [1998] 2 S.C.R. 217, at p. 257, “vouchsafes to the citizens and residents of the country a stable, predictable and ordered society in which to conduct their affairs.”  Given its fundamental importance in our society, it is not surprising that Parliament occasionally insists on taking some action to promote it, to safeguard it.  As was stated by Wilson J. in Operation Dismantle Inc. v. The Queen, [1985] 1 S.C.R. 441, at p. 489:  “There is no liberty without law and there is no law without some restriction of liberty:  see Dworkin, Taking Rights Seriously (1977), p. 267.”

116                           Permitting the exercise of the franchise by offenders incarcerated for serious offences undermines the rule of law and civic responsibility because such persons have demonstrated a great disrespect for the community in their committing serious crimes: such persons have attacked the stability and order within our community.  Society therefore may choose to curtail temporarily the availability of the vote to serious criminals both to punish those criminals and to insist that civic responsibility and respect for the rule of law, as goals worthy of pursuit, are prerequisites to democratic participation.  I say “goals worthy of pursuit” because it is clear that not all those who are otherwise eligible to vote are guaranteed to exercise civic responsibility, since, for example, there may be serious criminal offenders who may have avoided being apprehended and therefore still vote.  This does not, however, detract from the laudability of the goal.

117                           Related to the notion of the social contract is the importance of reinforcing the  significance of the relationship between individuals and their community when it comes to voting.  This special relationship is inherent in the fact that it is only “citizens” who are guaranteed the right to vote within s. 3  of the Charter .  This limitation of the scope of s. 3  of the Charter  stands in stark contrast to the protections offered by the fundamental freedoms, legal rights, and equality rights in the Charter , which are available to “everyone” or to “every individual”.  I am of the view that this limitation reflects the special relationship, characterized by entitlements and responsibilities, between citizens and their community.  It is this special relationship and its responsibilities which serious criminal offenders have assaulted.

118                           It is for this same reason, the importance of the nexus between voters and their community, that many jurisdictions qualify the right to vote with residency requirements.  This Court, in Haig v. Canada, [1993] 2 S.C.R. 995, upheld residency requirements as a reasonable qualification to the eligibility to vote in a referendum. While it is clear that there was no breach of s. 3  of the Charter  in that case since s. 3 does not apply to referenda, Haig, supra, generally seems to imply that residency requirements may be capable of being reasonable qualifications upon the right to vote.  This reasonableness arises not only from practical concerns, but also from the nexus between a particular individual’s eligibility to vote in an election, their relationship to the community, and the fact that it is that community which will be subjected to the results of the election.

119                           The American constitutional law scholar, Professor L. H. Tribe, notes in his text American Constitutional Law (2nd ed. 1988), at p. 1084:

Every state, as well as the federal government, imposes some restrictions on the franchise.  Although free and open participation in the electoral process lies at the core of democratic institutions, the need to confer the franchise on all who aspire to it is tempered by the recognition that completely unlimited voting could subvert the ideal of popular rule which democracy so ardently embraces. Moreover, in deciding who may and who may not vote in its elections, a community takes a crucial step in defining its identity. If nothing else, even though anyone in the world might have some interest in any given election’s outcome, a community should be empowered to exclude from its elections persons with no real nexus to the community as such. [Emphasis added.]

This analysis explains why citizenship or residency is a reasonable minimum requirement for voting, since such indicators are often equated with identification to a particular political community.  The importance of the nexus, however, also helps to understand the context of the particular disenfranchisement in question in the case at bar.  The disenfranchisement of serious criminal offenders serves to deliver a message to both the community and the offenders themselves that serious criminal activity will not be tolerated by the community.  In making such a choice, Parliament is projecting a view of Canadian society which Canadian society has of itself.  The commission of serious crimes gives rise to a temporary suspension of this nexus: on the physical level, this is reflected in incarceration and the deprivation of a range of liberties normally exercised by citizens and, at the symbolic level, this is reflected in temporary disenfranchisement.  The symbolic dimension is thus a further manifestation of community disapproval of the serious criminal conduct.

120                           From the perspective of persons whose criminal activity has resulted in their temporary disenfranchisement, their benefiting from society brought with it the responsibility to be subjected to the sanctions which the state decides will be attached to serious criminal activity such as they have chosen to undertake.  This understanding is complemented by the rehabilitative view that those who are in jail will hope and expect to regain the exercise of the vote on their release from incarceration, just like they hope and expect to regain the exercise of the fullest expressions of their liberty. Once released from prison, they are on the road to reintegration into the community. Obtaining the vote once released or paroled is a recognition of regaining the nexus with the community that was temporarily suspended during the incarceration.

121                           In Haig, supra, at p. 1031, L’Heureux-Dubé J. stated for the majority that “. . . in a democratic society, the right to vote as expressed in s. 3 must be given a content commensurate with those values embodied in a democratic state”.  I am of the view that the objectives pursued by Parliament in s. 51(e) of the Act do reflect values of the Canadian community, values related specifically to safeguarding the integrity and health of democracy itself.  These goals may be abstract, but they are the foundation of our society.  These are important values upon which s. 1  of the Charter  itself rests. A dimension of the rule of law is taking measures to see that law is obeyed.  Therefore, the disenfranchisement of serious criminal offenders is to be seen as protecting Canadian democracy rather than undermining it.  It is up to Parliament to enhance the value of the franchise.  It has responded with s. 51(e) of the Act, which is premised on the view that broadening of the franchise does not, in all cases, necessarily mean strengthening it.

G)     Prisoner Disenfranchisement in Canadian Provinces, Other Countries, and International Law: Illustration of a Range of Reasonable Balances

122                           As I noted above, a review of the legislation regarding prisoner disenfranchisement across Canadian provinces, in some other countries, and in some international instruments assists in demonstrating that there are a range of reasonable and rational balances that may be struck regarding this particular issue.

123                           In Canada, the situation across the provinces is obviously related to, and to some extent reflects, the evolution of the ongoing controversy regarding the federal legislation.  That having been said, an overview of provincial legislation actually demonstrates that a quite diverse range of balances has been struck with regard to provincial electoral law.  In Ontario, Quebec, Newfoundland, Prince Edward Island, and Manitoba, all prisoners vote:  Election Statute Law Amendment Act, 1998, S.O. 1998, c. 9, s. 13, repealing R.S.O. 1990, c. E.6, s. 16; Election Act, R.S.Q., c. E-3.3, s. 273; Elections Act, 1991, S.N. 1992, c. E‑3.1; Election Act, S.P.E.I. 1996, c. 12; Elections Act, R.S.M. 1987, c. E30, s. 31 (rep. & sub. S.M. 1998, c. 4, s. 21) disenfranchised “[e]very inmate of a correctional facility serving a sentence of five years or more”, but it was struck down by the Manitoba Court of Queen’s Bench in Driskell v. Manitoba (Attorney General), [1999] 11 W.W.R. 615.  In the Yukon, Saskatchewan and New Brunswick, all inmates serving sentences of imprisonment are disenfranchised:  Election Act, 1996, S.S. 1996, c. E-6.01, s. 17; Elections Act, R.S.N.B. 1973, c. E-3, s. 43(2)(e); and Elections Act, R.S.Y. 1986, c. 48, s. 5(d).  The Alberta Court of Appeal in Byatt v. Dykema (1998), 158 D.L.R. (4th) 644, struck down legislation providing for complete prisoner disenfranchisement, as it found itself bound by the decision of this Court in the first Sauvé case.  The Alberta Legislature then responded with legislation which provides for the disenfranchisement of all incarcerated persons serving a sentence of more than 10 days.  Section 45(c) of the Election Act, R.S.A. 2000, c. E-1, now declares ineligible to vote

persons who have been convicted of offences and on polling day are serving their sentences in a correctional institution under the Corrections Act, in a penitentiary under the Corrections and Conditional Release Act  (Canada) , in a place of custody under the Young Offenders Act  or the Young Offenders Act  (Canada)  or in any other similar institution outside Alberta, excluding persons sentenced to terms of imprisonment of 10 days or less or for the non-payment of fines.

The law in British Columbia, Nova Scotia, and the Northwest Territories and Nunavut parallels s. 51(e) of the Act that is being challenged in the case at bar: Election Act, R.S.B.C. 1996, c. 106, s. 30(b); Elections Act, R.S.N.S. 1989, c. 140, s. 29(d) as amended by S.N.S. 2001, c. 43, s. 13; Elections Act, R.S.N.W.T. 1988, c. E-2, s. 27(3), as amended by S.N.W.T. 1995, c. 14, s. 6.

124                           Turning to the United States, the U.S. Constitution does not explicitly protect the vote. In fact, § 2 of the Fourteenth Amendment makes tangential reference to the ability of the States to disenfranchise persons “. . . for participation in rebellion, or other crime”.  That having been said, the Fifteenth Amendment provides that the vote cannot be denied “on account of race, color, or previous conditions of servitude”.  Other constraints on the legislature’s ability to control the franchise include the Nineteenth Amendment, which disallows denial of the vote “on account of sex”; the Twenty-Fourth Amendment, which disallows denial “by reason of failure to pay any poll tax or other tax”; and the Twenty-Sixth Amendment, which disallows denial “on account of age” greater than 18 years.

125                           In the United States, it is the states that have control of disenfranchising inmates for both state and federal elections:  U.S. Constitution, Art. 1, § 2, cl. 1 (for the House of Representatives); Seventeenth Amendment, § 1 (for the Senate).  A general overview yields the conclusion that nearly all states (48 of 50 plus the District of Columbia) disqualify inmates incarcerated for felony offences for both state and federal elections, while some disenfranchise offenders permanently.  Only two states do not disqualify at all:  Maine and Vermont.  In November 2000, the Massachusetts electorate voted in favour of a State constitutional amendment limiting prisoners’ voting rights.

126                           A majority of the states which deprive inmates of the right to vote do so for the entirety of their sentence, including parole:  32 states prohibit felons from voting while they are on parole and 28 of those 32 also exclude felony probationers.  Other states allow a convicted felon to vote once incarceration ends.  Some states only remove the vote if the criminal has committed certain crimes.  A felon automatically regains the right to vote in most states upon completion of his or her sentence.  In a small number of states, a felon must apply for a pardon to be permitted to vote.  As mentioned above, some states remove the vote from convicted felons even after they have completed their sentences and paroles.  This practice was upheld as constitutional by the U.S. Supreme Court in Richardson v. Ramirez, 418 U.S. 24 (1974).

127                           Looking now to Europe, Art. 3 of the First Protocol to the ECHR (Eur. T.S. No. 9) guarantees “. . . free elections at reasonable intervals by secret ballot, under conditions which will ensure the free expression of the opinion of the people in the choice of the legislature”.  Article 3 has been considered on three occasions by the European Human Rights Commission (“the Commission”):  X. v. Netherlands, Application No. 6573/74, December 19, 1974, D.R. 1, p. 87; H. v. Netherlands, Application No. 9914/82, July 4, 1983, D.R. 33, p. 242; and Holland v. Ireland, Application No. 24827/94, April 14, 1998, D.R. 93-A, p. 15.

128                           In H. v. Netherlands, supra, the Commission established that Art. 3 of the First Protocol to the ECHR recognizes the principle of universal suffrage, but it also noted that the right to vote is not absolute and noted that “a large number of State Parties to the Convention have adopted legislation whereby the right to vote of a prisoner serving a term of imprisonment of a specific duration is suspended in certain cases, even beyond the duration of the sentence” (p. 245).  The general principle which supported the ability of the legislator to restrict the right to vote in respect of convicted persons was addressed as follows, at p. 246:

Such restrictions can be explained by the notion of dishonour that certain convictions carry with them for a specific period, which may be taken into consideration by legislation in respect of the exercise of political rights. Although, at first glance, it may seem inflexible that a prison sentence of more than one year should always result in the suspension of the exercise of the right to vote for three years, the Commission does not feel that such a measure goes beyond the restrictions justifiable in the context of Article 3 of the Protocol.

More recently, in Holland, supra, the Commission, at pp. 26-27, made a reference to:

. . . its constant case-law to the effect that, although Article 3 of Protocol No. 1 implies a recognition of the principle of universal suffrage (including the right to vote in elections for the legislature), this right is neither absolute nor without limitations but subject to such restrictions which are not arbitrary and which do not affect the expression of the opinion of the people in the choice of the legislature. . . .

In that same case, the Commission noted that it did not consider the disenfranchisement of prisoners for the duration of their incarceration to affect the expression of the opinion of the people in their choice of the legislature.

129                           The European Court of Human Rights addressed the issue in the Mathieu-Mohin and Clerfayt case, judgment of 2 March 1987, Series A No. 113.  Therein, the court found, at para. 51, that Art. 3 of the First Protocol conferred the right to vote and to stand for election, despite the wording of the Article, which, on its face, seems  not to confer such rights.  The court went on to state in para. 52 of its judgment, however, that:

The rights in question are not absolute.  Since Article 3 recognises them without setting them forth in express terms, let alone defining them, there is room for implied limitations . . . .   In their internal legal orders the Contracting States make the rights to vote and to stand for election subject to conditions which are not in principle precluded under Article 3 . . . .  They have a wide margin of appreciation in this sphere, but it is for the Court to determine in the last resort whether the requirements of Protocol No. 1 have been complied with; it has to satisfy itself that the conditions do not curtail the rights in question to such an extent as to impair their very essence and deprive them of their effectiveness; that they are imposed in pursuit of a legitimate aim; and that the means employed are not disproportionate . . . .  In particular, such conditions must not thwart “the free expression of the opinion of the people in the choice of the legislature”.

130                           European countries demonstrate a broad range of practices.  Seventeen European countries have no form of electoral ban for incarcerated offenders: Bosnia, Croatia, Cyprus, Denmark, Iceland, Ireland, Finland, Latvia, Lithuania, Macedonia, Netherlands, Poland, Slovenia, Spain, Sweden, Switzerland and the Ukraine.  In Greece, prisoners serving life sentences or indefinite sentences are disqualified; otherwise the matter is left to the discretion of the court.  In some other European countries, electoral disqualification depends on the crime committed or the length of the sentence:  Austria, Malta and San Marino ban all prisoners serving more than one year from voting; Belgium disqualifies all offenders serving sentences of four months or more; Italy disenfranchises based on the crime committed and/or the sentence length; Norway removes the vote for prisoners sentenced for specific offences; and in France and Germany, the disqualification of a prisoner is dependent upon the sentence handed down by the court specifically providing for disenfranchisement (in France, certain crimes are identified which carry automatic forfeiture of political rights; in Germany, prisoners convicted of offences which target the integrity of the German state or its democratic order lose the vote).  Armenia, Bulgaria, the Czech Republic, Estonia, Hungary, Luxembourg, Romania and Russia all have complete bans for sentenced offenders.

131                           Australia, New Zealand and the United Kingdom all disenfranchise at least some of the inmate population.  In Australia, prisoners vote in two of seven states.  In federal elections, inmates serving sentences of five years or more are disqualified from voting.  In New Zealand, prisoners serving sentences for three years or more, preventative detention or life imprisonment are not qualified to vote.  In the United Kingdom, prisoners are completely disenfranchised for Parliamentary elections, elections to the European Assembly, and local government elections:  s. 3(1) of the Represention of the People Act 1983 (U.K.), 1983, c. 2 (as amended in 1985 and 2000), and s. 2(1) of Schedule 1 to the European Assembly Elections Act 1978 (U.K.), 1978, c. 10.  The only exceptions are for remand prisoners, persons imprisoned for contempt of court and persons detained for default in complying with their sentence.

132                           When the Human Rights Act 1998 came into force in the United Kingdom in October 2000, the guarantee found in Art. 3 of Protocol No. 1 to the ECHR became part of U.K. law.  Three prisoners recently challenged the ban on prisoners’ voting as contained in the Representation of the People Act 1983, as being incompatible with the Human Rights Act 1998:  Pearson, supra.  In that case, the court construed the question before it as whether s. 3(1) of the Representation of the People Act 1983 satisfied the standard outlined by the European Court of Human Rights in Mathieu-Mohin, supra.  The analysis undertaken therein by Lord Justice Kennedy is informative, and very much in line with my disposal of the case at bar.  In fact, Lord Justice Kennedy devoted much of his analysis to the Canadian experience, and he specifically approved of the reasoning of Linden J.A. in the Federal Court of Appeal in the case at bar, with which I substantially agree.  Lord Justice Kennedy found that the balancing that was undertaken in the Canadian constitutional context was highly informative for the balancing which had to be undertaken under Art. 3 of Protocol No. 1 to the ECHR, as incorporated in the Human Rights Act 1998.  Lord Justice Kennedy stated, at para. 50, that:

. . . there is a broad spectrum of approaches among democratic societies, and the United Kingdom falls into the middle of the spectrum.  In course of time this position may move, either by way of further fine tuning, as was done recently in relation to remand prisoners and others, or more radically, but its position in the spectrum is plainly a matter for Parliament not for the courts.

I find this informative, especially in light of the fact that the line chosen by Parliament in Canada in s. 51(e) of the Act is far more moderate than the line drawn by Parliament in the United Kingdom.

133                           Certain international instruments also address the issue of prisoner voting. Article 25 of the International Covenant on Civil and Political Rights (“ICCPR”) states that every citizen shall have the “right and the opportunity” to vote “without unreasonable restrictions”:  ICCPR, 999 U.N.T.S. 171, entered into force March 23, 1976.  The United Nations Human Rights Committee, in a comment on Art. 25 of the ICCPR, stated that restrictions on the right to vote should be “objective and reasonable” and that “[i]f conviction for an offence is a basis for suspending the right to vote, the period of such suspension should be proportionate to the offence and the sentence”:  “General Comment Adopted by the Human Rights Committee under Article 40, Paragraph 4 of the International Covenant on Civil and Political Rights”, General Comment No. 25 (57), Annex V, CCPR/C/21, Rev. 1, Add. 7, August 27, 1996.  It is likely caveats, such as the one in Art. 25 of the  ICCPR, which led the international non-governmental organization Penal Reform International, in their 1995 publication Making Standards Work — An International Handbook on Good Prison Practice, at pp. 13-14, to distinguish between “retained rights”, which it advocated must be retained in a prison setting, and other rights which may be limited, amongst which was listed the right to vote.

134                           Therefore, when one looks to the range of balances selected by Canadian provinces, other countries, and as reflected in international instruments, it becomes clear that in theory there is not a single response to the question at hand.  The overview presents a range of reasonable and rational balances which have been struck. On the spectrum which is the result of the above overview, Canada’s line appears quite moderate.

H)  Application of the Oakes Test

135                           With all of the above commentary regarding flexibility, context, especially the relationship between the justification put forth by the Crown and a reasonable and logical prior social or political philosophy, the role of symbolic arguments, the importance of deference, and the variety of reasonable and rational balances struck by Canadian provinces and within the international community in mind, I now turn to the application of the Oakes test itself.

(1)  Is the Limit Prescribed by Law?

136                           It is obvious that the limit is prescribed by law: s. 51(e) of the Act is clear and not vague.

(2)  Are the Objectives Pressing and Substantial?

137                           The task of the Court at this phase of the Oakes test is to determine whether the concern which prompted the enactment of the impugned provision is pressing and substantial and whether the purpose of the legislation is one of sufficient importance: Irwin Toy, supra, at p. 987, per Dickson C.J., Lamer J. (as he then was) and Wilson J.  At this stage, there is an important distinction to be made between the objective and the means chosen to implement that objective, since this phase is related to the Court’s checking that the objectives are consistent with the principles, integral in a free and democratic society, pressing and substantial and directed to the realization of collective goals of fundamental importance:  Oakes, supra, at pp. 138-39.

138                           As a preliminary matter I think it is important to stress that, given the nature of the impugned provision and the nature of this particular context, it is advisable to not become over-constrained by the specific language or formulation of the objectives offered by the Crown.  Therefore, I agree with Linden J.A. in his general statement, at para. 99:

. . . 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. [Emphasis in original.]

139                           Parliament’s two principal objectives in s. 51(e) of the Act, accepted by both the trial judge and the Federal Court of Appeal below, are:  the enhancement of civic responsibility and respect for the rule of law, and the enhancement of the general purposes of the criminal sanction.  Above, I developed the view that these objectives are based upon a reasonable and rational social or political philosophy.  Thus, I am of the view that any provision which seeks to advance such objectives clearly has a pressing and substantial purpose.

140                           With regard to the first objective, that of enhancing civic responsibility and respect for the rule of law, the Crown submits that it basically relates to the promotion of good citizenship.  This objective also reflects society’s desire to buttress the rule of law.  I discussed both of these notions above, where I addressed how they relate to a reasonable and rational political or social philosophy, the view that the social rejection of serious crime reflects a moral line which safeguards the social contract and the rule of law and bolsters the importance of the nexus between individuals and the community.

141                           The Crown submits that it is illustrative to look to other jurisdictions so as to note that many liberal democracies do limit prisoner voting.  The trial judge, agreeing with Strayer J. in Belczowski (at the trial level), supra, questioned the usefulness of looking to other jurisdictions.  Particularly, after having surveyed the evidence, Wetston J. at trial below noted that it was difficult to draw meaningful conclusions about why some free and democratic societies disenfranchise criminal offenders while others do not.  It has been submitted that the fact that there is no single accepted theoretical basis for liberal democracy, no unified liberal political theory, means that civic responsibility should not be considered a significant objective of public policy. I do not agree with this submission.  I acknowledge that the practices of other liberal democracies, as reviewed above, are mixed.  I suggest, however, that while such evidence is clearly not capable of disposing of the issue, it is highly relevant.  The examination of other liberal democracies simply demonstrates that there is a range of reasonable and rational balances that have been struck.  The promotion of civic responsibility does not hinge on there being a single theory for liberal democracy.  The lack of there being a unified political theory is, so to speak, the point of the overview. Reasonable and rational persons and legislatures disagree on the issue of prisoner disenfranchisement.

142                           As noted above, it has been alleged that the “promotion of civic responsibility” is excessively abstract, generalized, symbolic, unrealistic and ambiguous.  I spent some time above discussing that symbolic or abstract purposes can be valid of their own accord and must not be downplayed simply for the reason of their being symbolic.

143                           The crux of the trial judge’s analysis regarding this first objective is found at pp. 882-83 of his reasons:

. . . 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.

144                           The pressing and substantial nature of this objective was agreed with by the majority of the Federal Court of Appeal below.  To support this, Linden J.A. cited as fn. 103 in para. 100 a passage from the decision of this Court in Harvey, supra, where the majority of this Court upheld similarly abstract objectives. At para. 38 of that case, La Forest J. wrote that:

I have no doubt that the primary goal of the impugned legislation is to maintain and enhance the integrity of the electoral process.  Nor do I doubt that such an objective is always of pressing and substantial concern in any society that purports to operate in accordance with the tenets of a free and democratic society. [Emphasis added.]

I am of the view that the objectives advanced in the case at bar are highly similar and are clearly pressing and substantial.

145                           Turning now to the second objective, that of the enhancement of the general purposes of the criminal sanction, the Crown submits that this objective has retributive, denunciatory and morally educative functions.  The appellants submit that this is an additional and gratuitous punishment, and serves only to comfort those outside prison by further stigmatizing those in prison as disconnected social outcasts.  They submit that a retributive philosophy of punishment does not correspond to the empirical realities of present-day Canadian society.

146                           Hugessen J.A., in Belczowski at the Federal Court of Appeal, and Arbour J.A. , in the first Sauvé case, were of the view that disenfranchisement of prisoners is problematic since it effectively amounts to a “punishment for imprisonment rather than for the commission of an offence”.  I disagree.  Disenfranchisement arising from s. 51(e) of the Act is directly related to the length of the sentence, reflecting the nature of the criminal offence and the criminal activity committed.  It is a valid objective for Parliament to develop appropriate sanctions and punishments for serious crime.  Further, it cannot be necessary for the nature of a sanction to be in some way directly related to the nature of the crime perpetrated, i.e. that only those convicted of electoral offences be deprived of the vote.  As discussed above, incarcerated persons lose the vote because they have been convicted of serious criminal offences and are the subjects of punishment.  The disenfranchisement is a civil disability arising from the criminal conviction.  The key fact which gives rise to the disenfranchisement is serious criminal activity as identified by Parliament.

147                           The trial judge held that s. 51(e) of the Act clearly does have a punitive aspect with a retributive function.  This conclusion was adopted by Linden J.A. at the Court of Appeal, in para. 100.

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.

148                           I am of the view that there was no error made by the courts below in identifying the objectives and in determining them to be pressing and substantial.  I think that the importance of both objectives is obvious.

(3)  Proportionality

(a) Rational Connection

149                           This first branch of the proportionality inquiry demands that this Court examine whether there is a rational connection between the disenfranchisement in s. 51(e) of the Act and the objectives of the legislation.  Therefore, the question is whether Parliament had a reasonable basis, based on the arguments and evidence advanced, for thinking that the temporary disenfranchisement of serious criminal offenders would augment civic responsibility and respect for the rule of law and enhance the general purposes of the criminal sanction. In Lavigne v. Ontario Public Service Employees Union, [1991] 2 S.C.R. 211, at p. 291, Wilson J. summarized what this inquiry demands:

The Oakes inquiry into “rational connection” between objectives and means to attain them requires nothing more than a showing that the legitimate and important goals of the legislature are logically furthered by the means government has chosen to adopt. [Emphasis added.]

150                           This Court has unanimously agreed that “[r]ational connection is to be established, upon a civil standard, through reason, logic or simply common sense”:  RJR-MacDonald, supra, per La Forest J., at para. 86, McLachlin J., at paras. 156-58, and Iacobucci J., at para. 184; referred to in Thomson Newspapers, supra, at para. 39. The existence of scientific proof is simply of probative value in supporting this reason, logic or common sense.  In the case at bar, as discussed above, a causal relationship between disenfranchising prisoners and the objectives approved of above is not empirically demonstrable.  However, this Court has clearly stated that Parliament must be afforded a margin of appreciation in regard to legitimate objectives which may, nonetheless, be based upon somewhat inconclusive social science evidence:  Sopinka J. in Butler, supra,  at pp. 502-3.  I offer two examples:  in Butler, Sopinka J. found, at p. 502, that it was “reasonable to presume” that there was a causal relationship between obscenity and harm to society, and in Ross v. New Brunswick School District No. 15, [1996] 1 S.C.R. 825, at para. 101, La Forest J. stated that it was “reasonable to anticipate” that there was a causal link between anti-Semitic activity by school teachers outside school and discriminatory attitudes within school.  Thus, it is clear that this Court’s approach to this dimension of the test demands not the strongest connection, the most convincing rational connection, but a logical or rational connection.

151                           In my view, it is obvious that s. 51(e) of the Act meets this requirement. The arguments in favour of the rational connection clearly meet the standard which this Court normally applies at this stage of the analysis, that the pressing and substantial legislative objectives are at least logically furthered by the means selected by Parliament in s. 51(e) of the Act.  At this phase of the proportionality inquiry, it is particularly important to continue to keep in mind the contextual factors which have been discussed above.

152                           The trial judge below stated that while he was of the view that there was a complete lack of empirical evidence to support the rational connection, this was not determinative, in that he was satisfied that there was a rational or common sense basis upon which to assume that disenfranchising prisoners is meant to promote civic responsibility.  He thus concluded that a rational connection was, as a matter of common sense, made out.  I agree.  He based this largely upon the view that the objectives here are symbolic:  society strongly disapproves of certain forms of conduct.  He also found the morally educative argument to be a compelling basis for the rational connection. The fact that many offenders are not caught, and thus perhaps those who are actually incarcerated are clearly singled out, also fails to undermine the rational connection.

153                           The crux of the trial judge’s reasoning concerning the first objective was that 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. 

(As per Linden J.A. at the Federal Court of Appeal, para. 105)

With regard to the second objective, the trial judge found at p. 892 of his reasons 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.

154                           The majority of the Federal Court of Appeal agreed with the trial judge that the promotion of civic responsibility is rationally connected to the impugned provision. The majority noted specifically that, while it was also possible to argue that all laws strive to promote civic responsibility, the common sense connection was adequately established in this case.  Linden J.A. also agreed with the trial judge that the objective of promoting the criminal sanction was also made out.

155                           Linden J.A. was satisfied that the two-year cutoff did catch serious and repeat offenders, based on the statistical data which showed that, as of April 1995, of a sample of 14,179 inmates in federal penitentiaries who had been sentenced to two or more years, each had committed, on average, 29.5 offences.  I think that this is highly relevant, particularly given the Chief Justice’s view that the disenfranchising provision is insufficiently tailored.  Wetston J., at pp. 878-79, looked to the criminal records of the appellants as illustrative of this point:

Mr. Sauvé, for example, was convicted of murder as an aider and abettor. While he committed only one significant offence, he was sentenced to a period of twenty-five years in prison.  In contrast, Mr. Spence’s criminal record reveals a history of repeated crimes which eventually led to a four-year term of imprisonment for armed robbery and related offences.  Clearly, both individuals engaged in serious criminal conduct, which the courts punished by way of prison sentences of more than two years.  Thus, the records of Mr. Sauvé and Mr. Spence also lend support to the defendants’ assertion that prison sentences of two years or more target serious criminals and repeat offenders.

156                           I note that in Harvey, supra, La Forest J. was convinced that the five-year disqualification from voting, holding office, or being elected to the Legislative Assembly for persons found guilty of illegal electoral practices found in the New Brunswick Elections Act was rationally connected to the legislative objective of enhancing the integrity of the electoral process.  He noted at para. 41:

A mandatory disqualification acts as a strong deterrent and helps to promote confidence in the electoral system . . . . [T]he . . . contention that the disqualification displays paternalism on the part of the legislature misses the point.  The provision is meant to protect the public not only from a particular offender, but from offenders in general.  In other words the legislature is aiming at both general and specific deterrence.

He concluded at para. 43 by noting:

. . . s. 119(c) is rationally connected to the objective of preserving the integrity of the electoral process and is not arbitrary in that it applies only to a specified group of individuals who are charged with and convicted of specified offences.

In my view, a similar analysis applies in the case at bar.

157                           I support the analysis of the courts below: reason, logic and common sense, as well as extensive expert evidence support a conclusion that there is a rational connection between disenfranchising offenders incarcerated for serious crimes and the objectives of promoting civic responsibility and the rule of law and the enhancement of the general objectives of the penal sanction.  The rational connection between the  disenfranchisement and the first objective is explained above, in my discussion of dignity and the fact that removing the right to vote from serious incarcerated criminals does no injury to, but rather recognizes their dignity (see paras. 68-76).  It is also explained above in the section entitled “A Rational and Reasonable Social or Political Philosophy Underpins the Crown’s Justification for the Limitation of the Section 3 Right” (see especially paras. 114-121), and below in my discussion of the salutary effects of the measure (see especially paras. 180-183).  In the latter section, I discuss the legislation’s expression of societal values and its signalling effect.  The Chief Justice prefers a different line of reasoning.  Citing Mill as her authority, she states that “denying penitentiary inmates the right to vote is more likely to send messages that undermine respect for the law and democracy than messages that enhance those values” (para. 41).  However, apart from one philosopher, she provides no support for this contention; she simply replaces one reasonable position with another, dismissing the government’s position as “unhelpful” (para. 37 of the Chief Justice’s reasons).

158                           The rational connection between the legislation and the enhancement of the criminal sanction is also elaborated on elsewhere.  Below, at paras. 160 to 174 on minimal impairment, I explain at length that the disenfranchisement is carefully tailored to apply to perpetrators of serious crimes.  I therefore disagree with the Chief Justice’s statement that denial of the right to vote is insufficiently tailored and therefore not rationally connected to legitimate punishment.  I also explain below, at para. 183 of  my discussion of salutary effects, that denial of the right to vote is perceived as meaningful by the prisoners themselves (the evidence of the appellant Aaron Spence supports this) and can therefore contribute to the rehabilitation of prisoners.

159                           The arguments raised by the appellants are not destructive of the rational connection between Parliament’s objectives and legislative action, and do not support disregarding the findings of both lower courts.  As per Wilson J. in Lavigne, supra, Parliament’s goals are obviously “logically furthered by the means government has chosen to adopt” (p. 291 (emphasis added)).  In particular, I share the view of the courts below that given that, the objectives are largely symbolic, common sense dictates that social condemnation of criminal activity and a desire to promote civic responsibility are reflected in disenfranchisement of those who have committed serious crimes.  This justification is rooted in a reasonable and rational social and political philosophy which has been adopted by Parliament.  Further, it can hardly be seen as “novel”, as stated in the Chief Justice’s reasons, at para. 41.  The view of the courts below is that generally supported by democratic countries.  Countries including the United States, the United Kingdom, Australia, New Zealand, and many European countries such as France and Germany, have, by virtue of choosing some form of prisoner disenfranchisement, also identified a connection between objectives similar to those advanced in the case at bar and the means of prisoner disenfranchisement.

(b)  Minimal Impairment

160                           The Crown must demonstrate that the impairment of rights is minimal, i.e. that the law was carefully tailored so that Charter  rights are impaired no more than is necessary to meet the legislative provision’s objectives.  Minimal impairment is about analysing the line that has been drawn.  This analysis does not, notably, require the Crown to have adopted the absolutely least intrusive means for promoting the purpose, although it does require that the Crown prefer a significantly less intrusive means if it is of equal effectiveness.  In RJR-MacDonald, supra, at para. 160, McLachlin J. 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. . . .

In R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 713, at p. 782, Dickson C.J.  stated:  “The courts are not called upon to substitute judicial opinions for legislative ones as to the place at which to draw a precise line.”  See also Ross, supra, at para. 108.  In the case at bar, these passages are of the utmost relevance, insofar as they encompass a recognition of the need for deference in the context of a case such as the one at bar.

161                           I emphasize that it was “particularly” on the ground of minimal impairment that this Court, in the first Sauvé case, established that the previous s. 51(e) of the Act, which disenfranchised all prisoners regardless of the duration of their incarceration, was contrary to the Charter  and incapable of being justified under s. 1.  Our decision was, at pp. 439-40:

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.

The language of Iacobucci J.’s reasons seem to imply that, while Parliament’s complete ban of prisoner voting in the old provision was unconstitutional, Parliament was free to investigate where an appropriate line could be drawn.  This is exactly what it was in the process of doing at the time the first Sauvé case was heard.  It has drawn a line in the form of s. 51(e) of the Act.

162                           The appellants and their experts have argued that there are less intrusive means for the Crown to pursue its objectives: disenfranchisement could be left to the discretion of the sentencing judge; as per the Lortie Commission, only those convicted of the most serious offences (those punishable by a maximum of life) and the most serious sentences (those punishable by 10 years in jail or more) could lose the vote; an offence-oriented approach could define specific types of crimes which could be seen as bearing a rational connection to the franchise; or the measure could allow for the vote to be restored if the offender demonstrated good behaviour while incarcerated.  To these I add that it is obvious that any higher cutoff line, i.e. 5, 10, or 25 years of incarceration, would also, technically, be less intrusive.

163                           I am of the view that no less intrusive measure would be equally effective. Since Parliament has drawn a line which identifies which incarcerated offenders have committed serious enough crimes to warrant being deprived of the vote, any alternative line will not be of equal effectiveness.  Equal effectiveness is a dimension of the analysis that should not be underemphasized, as it relates directly to Parliament’s ability to pursue its legitimate objectives effectively.  Any other line insisted upon amounts to second-guessing Parliament as to what amounts to “serious” crime.

164                           The trial judge below stressed that the legislative process did undertake a rigorous evaluation of the line chosen regarding “serious” crime.  He noted that the legislative history of s. 51(e) of the Act involved a report from the Lortie Commission.  Notably, the Lortie Commission, despite the conclusion of a Research Study commissioned under it which recommended that all prisoners get the right to vote, concluded that prisoners who had been convicted of an offence punishable by a maximum of life imprisonment and who had been sentenced to a prison term of 10 years or more should be disqualified from voting for the duration of their incarceration.  A Special Committee on Electoral Reform, which reviewed the Lortie Commission’s Report, recommended, however, that a two-year cutoff was appropriate since this would catch “serious offenders”.  Wetston J. stated, at p. 877:

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.

165                           The debates in the House of Commons regarding the provision in question demonstrate that there was a view that disenfranchisement would have an “educative effect”:  see Wetston J. at pp. 877-78 of his reasons and House of Commons Debates, vol. XIV, 3rd Sess., 34th Parl., April 2, 1993, at pp. 18015-21.  Further, the debates emphasize the view that the prisoners actually disenfranchised themselves by their engaging in criminal conduct.  Most generally, and most importantly, Linden J.A. aptly noted, at para. 96:

. . . 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.

Linden J.A. noted that Parliament did not consider a finite list of offences which would result in the loss of the vote, but he noted that the cutoff selected achieved the same objective by a different method: serious offences were specifically targeted.

166                           The Crown and its experts submitted that the impugned provision is minimally impairing for three reasons:  only prisoners serving sentences for two years or more are disenfranchised, and thus the provision only targets what Parliament has identified as those who have perpetrated “serious” offences; the disenfranchisement is temporary, in the sense that the vote returns to the offenders once they leave jail; and the return of the vote once the offender leaves jail is automatic.

167                           The trial judge, appropriately, was concerned that Parliament be granted some latitude to select alternative policy options.  In his final analysis, however, Wetston J. rejected all but one of the appellants’ suggested “less intrusive” options.  What he believed would be a less intrusive measure is the selective disenfranchisement of each individual offender as a matter of the sentencing judge’s discretion.  He preferred this approach since  it would not be automatic; it would take into account the particular circumstances of each offender.  He noted that this option was not heavily considered by Parliament.  Further, he noted that legislative criteria could assist the sentencing judge in his or her determinations.  The trial judge was also of the view that there would be a greater degree of public attention if the issue of disqualification were to be decided by the sentencing judge, and thus the morally educative dimension of the disenfranchisement would be enhanced.  The trial judge thus found that in light of this available option the provision was not minimally impairing.  Desjardins J.A., in dissent at the Federal Court of Appeal, was effectively of the same view.

168                           I am not convinced by the trial judge’s preference for this approach.  First, it seems to designate the actual purpose of the voting ban to be more clearly the enhancement of the punitive sanction than the promotion of civic responsibility.  I am of the view that if the offender has committed a crime which falls within the category identified by Parliament as bringing with it disenfranchisement, the sentencing judge’s discretion has little place in determining whether civic responsibility will be better or worse promoted by the disenfranchisement of the particular offender.  It is therefore clearly not of equal effectiveness.  Second, I cannot imagine what factors a sentencing judge would appropriately weigh to decide this issue.  The “personal circumstances” which a sentencing judge might employ to reduce a sentence have, in my view, no place in determining whether a particular offender be permitted to vote.  The disenfranchisement comes solely from  having committed a crime serious enough to bring with it a term of incarceration of two years or more.

169                           The trial judge also noted that legislative guidelines could be provided to assist the sentencing judge.  I fail to see how such guidelines would be any more effective than the line currently drawn since the current line is Parliament’s choice of a guideline:  it clearly identifies a line which Parliament believes represents what amounts to “serious crime”.  The suggestion that a sentencing judge be able to exercise a discretion, in my view, trivialises the gravity of all offences which Parliament has said, by virtue of the cutoff line selected in relation to the Criminal Code , are serious.

170                           It could be said that there are many other less intrusive methods to promote civic responsibility or the rule of law, methods which do not involve prisoners at all.  For example, it could be said that a program of compulsory voting for all citizens, as is in effect in Australia or Belgium, might be such an alternative.  With regard to such suggestions, I note the following passage from Dickson C.J. in Keegstra, supra, at pp. 784-85:

. . . s. 1 should not operate in every instance so as to force the government to rely upon only the mode of intervention least intrusive of a Charter  right or freedom. It may be that a number of courses of action are available in the furtherance of a pressing and substantial objective, each imposing a varying degree of restriction upon a right or freedom.  In such circumstances, the government may legitimately employ a more restrictive measure, either alone or as part of a larger programme of action, if that measure is not redundant, furthering the objective in ways that alternative responses could not, and is in all other respects proportionate to a valid s. 1 aim.  [Emphasis added.]

171                           The approach taken by Linden J.A. for the majority of the Federal Court of Appeal below is sound.  He held that the current provision was minimally impairing for a number of reasons:  only “serious offenders”, as determined by Parliament, are subject to disenfranchisement; accused persons and those convicted but out on bail are permitted to vote; those out on parole are permitted to vote; and the provision acts proportionately since it is individualized insofar as those with longer sentences will be disenfranchised longer.  The provision is reasonably tailored insofar as disenfranchisement reflects the length of the sentence and actual incarceration, which, in turn, reflects the seriousness of the crime perpetrated and the intended progress towards the ultimate goals of rehabilitation and reintegration.

172                           Linden J.A. also, correctly in my view, drew a parallel to the approach to minimal impairment adopted by this Court in Harvey, supra.  He noted at para. 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.

173                           I agree with the Crown that the impugned provision is not arbitrary:  it is related directly to particular categories of conduct.  I also note that the two-year cutoff line also reflects practical considerations:  it reflects a distinguishing between offenders incarcerated in federal rather than provincial institutions (s. 743.1 of the Criminal Code ); persons sentenced to a term of imprisonment of two years or more are not eligible to serve their sentence in open custody (s. 742.1  of the Criminal Code ); and persons subject to a sentence of imprisonment of two years or more are subject to having a court delay parole until one half of the sentence is served (s. 743.6  of the Criminal Code ).

174                           In my view, it is particularly inappropriate, in the case at bar, to find the justification of the limitation of the right to be unconvincing at this  phase of the Oakes test.  First, as was noted above, there is a need for deference to Parliament in its drawing of a line, especially since this Court gave the impression that it was up to Parliament to do exactly that after the first Sauvé case was heard in 1993.  Second, also as developed above, the analysis of social and political philosophies and the accommodation of values in the context of the Charter  must be sensitive to the fact that there may be many possible reasonable and rational balances.  Developing this point, it is important to note that, given the theoretical nature of the arguments raised by both parties in the case at bar, they do not gain proportionally in strength as the bar is moved higher.  Symbolic and theoretical justifications such as employed in this case do not get stronger as the line changes.  The fundamental premises underlying the line chosen would be the same if the cutoff was 10 years, or even 25 years.  See, for example, Driskell, supra, in which similar analytical problems to those in the case at bar arose and resulted in a line of five years being held unconstitutional.  Line drawing, amongst a range of acceptable alternatives, is for Parliament.  This view is compounded by the point developed above that it is plain that any alternate line would not be equally effective, in that the line drawn reflects Parliament’s identification of what amounts to serious criminal activity.  The Federal Court of Appeal was correct to find the provision in question minimally impairing.

(c)  Proportionality of Effects

175                           The final prong of the Oakes test demands that the effects of the limiting measure (the impugned provision) must not so severely trench on Charter  rights that the legislative objective, albeit important, is outweighed by the infringement of the rights. The basic test for determining proportionality is that the objectives must be balanced with the actual effects of the impugned provision:  Oakes, supra; Edwards Books, supra; McKinney v. University of Guelph, [1990] 3 S.C.R. 229.  This basic test, however, was restated and modified by Lamer C.J. in Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835.  Therein, it was held that in cases where a measure fully, or nearly fully, meets its legislative objective or objectives, then the conventional Oakes analysis stands:  weigh the objectives with the actual effects of the impugned provision.  Where a measure only partially achieves its legislative objective or objectives, the proportionality requirement is dual: not only must there be proportionality between the deleterious effects of the measure which are responsible for the limiting of the right in question and the objective, but there must also be proportionality between the deleterious and the salutary effects of the measures.

176                           What of this case?  The Crown alleges that the objectives have been fully met, and thus the pre-Dagenais approach should apply.  As submitted by the Crown at para. 64 of their factum, “[t]he enactment of the legislation creates the norm and fulfills the moral aim of the legislation. As in Harvey . . . there is no need to balance the salutary and deleterious effects of this legislation”.

177                           In Harvey, supra, at para. 48, La Forest J. stated that “[t]he final step in the Oakes analysis is to determine if the effects of s. 119(c), the removal of the appellant as the member for Carleton North and his five-year disqualification from running as a candidate, are proportional to the section’s objective of ensuring the integrity of the electoral process.”  Thus, La Forest J. did not go on to balance salutary and deleterious effects per se; the emphasis was only on weighing the proportionality of the deleterious effects to the objectives of the provision.  While Linden J.A. did not definitively prefer the Harvey approach in this case, he noted at para. 133 that it “highlights that the context of the particular case is paramount in the Oakes analysis”.  Further, he noted at para. 134 that “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”.  I agree and am of the view that regardless of which test is engaged, given the nature of the evidence and the fact that the objectives have clear symbolic effect, that the proportionality analysis is nonetheless satisfied.

178                           It is my view that the arguments in this dimension of the analysis are basically either persuasive or not. If the objectives are taken to reflect a moral choice by Parliament which has great symbolic importance and effect and which are based on a reasonable social or political philosophy, then their resulting weight is great indeed.  Over all, while the temporary disenfranchisement is clear, the salutary effects and objectives are, in my view, of greater countervailing weight.  Generally, I agree with the analysis of Linden J.A. at the Federal Court of Appeal below to this effect.

179                           The trial judge considered this dimension of the Oakes test despite having found that the impugned provision was not minimally impairing.  He discussed the current situation across Canadian provinces with regard to prisoner enfranchisement for the purpose of provincial elections.  He noted that four provinces (I note that it is now five) permit all prisoners to vote in provincial elections, others place some limits, while yet others provide for complete disenfranchisement.  He then found that the Crown did not provide any evidence of harm flowing from instances where prisoners had exercised the right to vote, such as provincial elections or referenda.  He also noted that the Crown did not provide any evidence of harm flowing from prisoner voting in other countries. I do not find this reasoning persuasive: the harm which flows from serious offenders voting is obviously not empirically demonstrable.  As long as one holds democracy to be an abstract good, to find that empirically measurable harm flows from the result of any fair democratic process is an impossible argument to make.

180                           The salutary effects in the case at bar are particularly difficult to demonstrate by empirical evidence given their largely symbolic nature.  On this point, I note that it would be difficult for the Crown to justify all penal sanctions, if scientific proof was the standard which was required.  I discussed this above, and would like to reiterate that many core values of the Canadian community might suffer if put to such a test.  In such cases, the weighty merit of the objectives themselves must be considered with the social, legislative and factual context in mind. In this case, a central dimension of the context is Parliament’s choice of a particular social or political philosophy on which the justification for the limitation of the right is based.  As Bastarache J. noted in Thomson Newspapers, supra, at para. 125, this third phase of the proportionality prong of the Oakes test is unique in that it

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 .

181                           Linden J.A. found that the primary salutary effect was that the legislation, intrinsically, expresses societal values in relation to serious criminal behaviour and the right to vote in our society.  He thus concluded, at para. 137, that it has more than symbolic effect:

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.  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”.  [Footnotes omitted; emphasis in original.]

Linden J.A. suggested that value emerges from the signal or message that those who commit serious crimes will temporarily lose one aspect of the political equality of citizens.  Therefore,  “the enactment of the measure is itself a salutary effect” (para. 138).  I agree.  As can be drawn out from the overview of the arguments which were placed before this Court, one is forced to either accept the objectives, and consequently grant them weight at this stage of the analysis, or discount them.  I am of the view that the salutary effects and objectives must be granted the respect of this Court.

182                           The signalling function of s. 51(e) is highly important.  Linden J.A., below, noted at paras. 139 and 141 of his reasons, that:

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.

. . .

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. [Footnotes omitted.]

This view is supported by the Crown’s experts, Professors Pangle and Manfredi. Professor Pangle emphasized that the provision in question reflects a signal to society and to offenders that the commission of serious criminal acts will result in the loss of a dimension of political participation.  Professor Manfredi developed the argument, employed above, that there is something important about the nexus between the exercise of the vote by citizens and the responsibilities and duties inherent in citizenship.  Also as reflected upon above, the provision reflects society’s desire to promote the rule of law and civic responsibility, which are inherent in and required by the rule of law and the notion of the social contract.  This provision draws a line which sends a strong and beneficial moral message to society as a whole, the message that crime will not be tolerated.  As was said in Sharpe, supra, by L’Heureux-Dubé and  Bastarache JJ. and myself, at para. 191: “The Court should be particularly sensitive to the legitimate role of government in legislating with respect to our social values”.

183                           Another salutary effect of the impugned provision is that it has an effect that is noticed by the disenfranchised offenders themselves.  Linden J.A. agreed and substantiated this view by quoting testimony from the appellant Aaron Spence to the effect that Mr. Spence was troubled by the fact that he was denied the vote, at para. 143:

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.

I am of the view that this evidences the salutary effect that the temporary disenfranchisement is perceived as meaningful by the offenders themselves.  I am also of the view that based on this, there is reason to believe that the disenfranchisement could have an ongoing positive rehabilitative effect.  Since the vote is meaningful to these offenders, then perhaps its temporary loss will be a factor which these offenders will carry with them as they pursue reintegration into the community on their release.  The reasons of the Chief Justice deny this and suggest that the denial of the vote to prisoners makes reintegration and rehabilitation more difficult.  As just stated, I disagree.  I note as well, however, that it is possible to argue that incarceration itself may make rehabilitation and reintegration more difficult, but it is still, in some cases, an important dimension of punishment and indeed a step towards rehabilitation.

184                           The most obvious deleterious effect of s. 51(e) is the potential temporary loss of the vote.  This, however, must be considered in light of Parliament’s objectives, as illuminated by the totality of the context.  Based on statistical data, Linden J.A. concluded that the provision was effective insofar as it affected only the most serious offenders.  Linden J.A. presented an overview of the number and percentage of inmates in federal penitentiaries serving time for particular offences in the form of a table reproduced at para. 145 of his reasons.  Basically, it reflected the serious nature of the offences committed by those persons incarcerated for two years or more.  As noted above, of the sample of 14,179 offenders incarcerated for two years or more, each was found to have, on average, 29.5 convictions.  Linden J.A. thus properly concluded that the provision catches serious and repeat offenders.  The statistics also indicate that 75 percent of prisoners incarcerated in federal penitentiaries are serving sentences of five years or less.  Therefore, most prisoners will only be deprived of participation in one election: see Linden J.A., at para. 145;  see also Linden J.A., at para. 122, where fixed periods of disenfranchisement are discussed and a helpful parallel is drawn to this Court’s approach in Harvey, supra.  Lastly, because the duration of the disenfranchisement is directly related to the duration of incarceration, it may in fact be the case that a serious criminal offender who is technically disenfranchised during the period of his or her incarceration may never actually be denied the opportunity to vote, as there may be no election during the time he or she is incarcerated.

185                           The reasons of the Chief Justice suggest that to be temporarily disenfranchised while incarcerated is to be severed from the body politic and silenced as an unworthy outsider.  Above, I explained how temporary disenfranchisement does not undermine the “worth” or “dignity” of any offender but is instead focussed at criminal offences.  I also have discussed how temporary disenfranchisement is to be seen as a dimension of punishment that is tailored towards rehabilitation and reintegration:  it is therefore ultimately focussed upon inclusion rather than exclusion.  One other point which I would like to make briefly is to note that, while being temporarily disenfranchised is clearly a significant measure, which is part of the reason why it carries such great symbolic weight, it does not amount to the complete extinguishment of all means of political expression or participation.  There are many other avenues by which serious criminals who are incarcerated for two or more years may still exercise political expression:  they can write to and lobby elected representatives, publish their ideas or policy proposals, or in other ways make their views known.

186                           I return to the issue of deference to Parliament, and to the special context of this case, that the justification advanced by Parliament is rooted in a social or political philosophy that is not susceptible to proof in the traditional sense.  Linden J.A. noted at the Federal Court of Appeal, at para. 114, that:

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.  Such considerations are by definition political and therefore warrant deference.

Deference is appropriate since the impugned provision raises questions of penal philosophy and policy.  Linden J.A. stated, at para. 135 that “[t]he courts cannot prevent Parliament from proportionately compromising Charter  rights in the name of denouncing crime, even if they disagree with Parliament’s penal philosophy”.  On the issue of deference to choices regarding penal philosophy, I refer to this Court’s decision in R. v. Goltz, [1991] 3 S.C.R. 485, where it upheld a mandatory sentence imposed under the British Columbia Motor Vehicle Act, on the basis that it was not an infringement of the guarantee against cruel or unusual punishment found in s. 12  of the Charter .  In Goltz at p. 502, speaking for a majority of the Court, I cited from R. v. Guiller (1986), 48 C.R. (3d) 226 (Ont. Dist. Ct.), per Borins J., the following passage, approved of by Lamer J. in both R. v. Luxton, [1990] 2 S.C.R. 711, at p. 725, and R. v. Smith, [1987] 1 S.C.R. 1045, at p. 1070:

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. . . .

187                           In his reasons, Linden J.A. made the appropriate comment that the recent evolution of penal policy has been an issue attracting significant political attention.  At para. 116 he stated:

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.

188                           When the objectives and the salutary effects are viewed in the totality of the context, they outweigh the temporary disenfranchisement of the serious criminal offender which mirrors the fact of his or her incarceration.  In my view, Parliament has enacted a law which is reasonable, and which is justified in a free and democratic society.

V.  Section 15(1)  of the Charter 

189                           I agree with the trial judge and unanimous Court of Appeal below (Desjardins J.A. was in agreement with Linden J.A.’s reasoning regarding s. 15(1)) that it is clear that there has been no infringement of s. 15(1)  of the Charter .

190                           In Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497, Iacobucci J. summarized, at para. 88, the proper approach to s. 15(1) as follows:

(3)  . . . a court that is called upon to determine a discrimination claim under s. 15(1) should make the following three broad inquiries:

(A)      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?

(B)       Is the claimant subject to differential treatment based on one or more enumerated and analogous grounds?

and

(C)      Does the differential treatment discriminate, by imposing a burden upon or withholding a benefit from the claimant in a manner which reflects the stereotypical application of presumed group or personal characteristics, or which otherwise has the effect of perpetuating or promoting the view that the individual is less capable or worthy of recognition or value as a human being or as a member of Canadian society, equally deserving of concern, respect, and consideration?

191                           In M. v. H., [1999] 2 S.C.R. 3, at para. 63, Cory and Iacobucci JJ. noted that for distinctions to be discriminatory, they must be made on the basis of an enumerated or analogous ground:

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.]

192                           Even if I were to presume that a distinction has been based on personal characteristics pursuant to inquiry (A) of the Law criteria, I am of the view that the answer to (B) is clearly in the negative.  The status of being a prisoner does not constitute an analogous ground.

193                           Several courts have canvassed the issue of whether prisoner status constitutes an analogous ground: Jackson v. Joyceville Penitentiary, [1990] 3 F.C. 55 (T.D.); Belczowski (at the trial level), supra; McKinnon v. M.N.R., 91 D.T.C. 1002 (T.C.C.); Armstrong v. R., [1996] 1 C.T.C. 2745 (T.C.C.); Mulligan v. R., [1997] 2 C.T.C. 2062 (T.C.C.), followed in Wells v. R., [1998] 1 C.T.C. 2118 (T.C.C.); Olson v. Canada, [1996] 2 F.C. 168 (T.D.), leave to appeal refused, [1997] 3 S.C.R. xii; and Alcorn v. Canada  (Commissioner of Corrections) (1999), 163 F.T.R. 1, aff’d (2002), 95 C.R.R. (2d) 326, 2002 FCA 154.  All have held that prisoner status does not amount to a ground analogous to those enumerated under s. 15(1)  of the Charter .

194                           In Chiarelli v. Canada (Minister of Employment and Immigration), [1992] 1 S.C.R. 711, this Court agreed with the reasoning of the Federal Court of Appeal below on the point that permanent residents who had been convicted of criminal offences involving terms of imprisonment of five years or more did not constitute a category of persons analogous to those enumerated in s. 15(1).  Pratte J.A. at the Federal Court of Appeal, [1990] 2 F.C. 299, at p. 311, stated:

No analogy can be made between the grounds of discrimination mentioned in section 15 and the fact that certain permanent residents have been convicted of serious offences.  Permanent residents who have been convicted of serious criminal offences do not fall into an analogous category to those specifically enumerated in section 15.

I am of the view that a similar analysis applies in the case at bar, although here the distinction drawn is between citizens who have committed serious criminal offences for which they have been incarcerated as punishment and other citizens, rather than the distinction in Chiarelli drawn between permanent residents who have committed serious offences warranting terms of imprisonment of five years or more and other permanent residents.

195                           Prisoners do not constitute a group analogous to those enumerated in s. 15(1) because the fact of being incarcerated cannot be said to have arisen because of a stereotypical application of a presumed group characteristic.  The status of being a prisoner is brought about by the past commission of serious criminal offences, acts committed by the individual himself or herself.  The unifying group characteristic is past criminal behaviour. This was the view of the trial judge in Jackson, supra, noted by the trial judge in the case at bar: the differential treatment arises “not from their personal characteristics, but from past courses of conduct” (p. 920).  This was also the view of Strayer J. in Belczowski at the trial level, supra, at p. 162:  “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)”.

196                           Linden J.A., below, correctly stated at para. 166 of his reasons:

. . . 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.

197                           With regard to the argument that being imprisoned could be said to be immutable or constructively immutable, insofar as incarceration is, obviously, once it has begun, beyond a prisoner’s control, the immutability or constructive immutability is nonetheless an inherent and necessary dimension of being incarcerated, which obviously and validly relates to the state’s legitimate role of punishing serious criminals for their criminal activity.  If one could change the fact of being incarcerated at one’s own whim, then it would be a useless concept.  Most importantly, while being incarcerated is beyond the immediate control of a prisoner, it is the result of their criminal activity, the commission of which was within their control.  As Professor Hogg has aptly suggested, at p. 52-29 of Constitutional Law of Canada, supra:

Another way of looking at immutability as the common element of the listed personal characteristics is to notice that the characteristics are inherent, rather than acquired.  They do not reflect a voluntary choice by anyone, but rather an involuntary inheritance. . . . Section 15 prohibits laws that distinguish between people on the basis of their inherent attributes as opposed to their behaviour.  Section 15 therefore does not prohibit laws that make special provision for those who have committed a crime, become insolvent, manufactured food or drugs, joined the legal profession, made a will, purchased a taxable good or service, etc.  It is true that individuals may claim to be treated unfairly by the law for conditions that are their own responsibility, but this kind of claim even if fully justified does not warrant a constitutional remedy.

198                           When discussing punishment and incarceration, it is important to also address s. 12  of the Charter .  Section 12 , which protects “[e]veryone” from “cruel and unusual treatment or punishment”, addresses the condition of incarcerated persons and persons undergoing other forms of punishment or treatment.  It, however, inherently recognizes the fact that incarcerated persons are clearly subject to a kind of unequal treatment, but a form of treatment that is nonetheless constitutional as long as it is not “cruel and unusual”.

199                           I spent much time above addressing how being incarcerated for having committed a serious crime does not go to the “dignity” or “worth” of the perpetrator:  it relates to the crime committed.  Crime is appropriately disapproved of:  disapproval is inherent in its definition, which relates to committing offences which are punishable by law.  Prisoner status relates to the valid fact of punishment for this criminal activity.

200                           Further, from the perspective of the general community and of victims or potential victims, I note that serious criminal conduct is often directly inimical to one of the two purposes of s. 15(1), that of ensuring the equal worth and dignity of all human persons:  see Eldridge v. British Columbia (Attorney General), [1997] 3 S.C.R. 624, at para. 54.

201                           It is clear that the case law of this Court indicates that equality does not necessarily connote identical treatment.  In my view, to find prisoner status to be an analogous ground would be a distortion of the purpose of s. 15(1) and would come close to making a mockery of the Criminal Code  and the values on which it is based and which it enshrines.  This is because it makes little sense to suggest that the distinction being drawn between incarcerated persons and other citizens is based on a stereotypical or irrelevant ground.  The relevant and valid ground of differentiation which results in becoming a prisoner is the conviction for a serious criminal offence.

202                           An alternative argument was made before this Court that imprisonment should be recognized as an analogous ground because of adverse effect or impact discrimination based on the fact that Aboriginal peoples make up a “disproportionate” percentage of prisoners.  I am not convinced by this argument.  On this point, I adopt the following analysis of Linden J.A., from para. 169 of his reasons:

First, since, according to the data offered, 1,837 Aboriginal people are disenfranchised by this law, 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.  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.  [Footnotes omitted.]

Beyond this, I find any analysis of adverse impact or effect discrimination seems parasitic on finding that prisoner status constitutes an analogous ground, since the adverse effect or impact is allegedly upon Aboriginal prisoners; it is the prisoner status which is alleged to result in the disadvantage, something which is not furthered within the category of Aboriginal prisoner status.  It is clear that if the treatment was targeted against Aboriginal people in a direct way, they would have a valid objection based on an enumerated ground of s. 15(1).  As Linden J.A. noted, the numbers do not warrant the conclusion that the disenfranchisement provision is being used to target Aboriginal persons per se or has this effect.

203                           The key is that the fact of incarceration does not necessarily arise due to any personal attribute such as race or ethnic origin and neither does it necessarily relate to social condition.  It necessarily relates to having committed crimes.  If Aboriginal inmates had their votes taken away for life, while non-Aboriginal inmates only suffered a temporary suspension, or if Aboriginal inmates were disenfranchised but not non-Aboriginal inmates, then such differential treatment would clearly warrant different analysis.  But this is not the case here.

204                           The reasons of the Chief Justice, at para. 60, refer to the fact that this Court, in R. v. Gladue, [1999] 1 S.C.R. 688, noted that the overrepresentation of Aboriginal persons in the criminal justice system and the prison population reflects a “crisis in the Canadian criminal justice system”.  I agree that a sad and pressing social problem exists, but suggest that it is quite a leap to then say that Parliament is incapable of enacting a provision which disenfranchises serious criminal offenders who have been sentenced to two or more years of incarceration.  As noted above, it is not plausible to say that the temporary disenfranchisement is in some way targeted at Aboriginal people:  it hinges only upon the commission of serious criminal offences.  If there is a problem with the overrepresentation of Aboriginal people in our criminal justice system and prisons, then that issue must continue to be addressed, by not only continuing to pay attention to the sentencing considerations pursuant to s. 718.2 (e) of the Criminal Code , which are specifically aimed at such a reduction, but also by addressing some of the root causes of the overrepresentation identified by this Court in Gladue, supra, including poverty, substance abuse, lack of education, lack of employment opportunities, and bias against Aboriginal people.  The continuing need to address these factors does not, however, preclude the ability of Parliament to address other pressing social problems, including denouncing serious crime, enhancing the meaning of the criminal sanction and promoting civic responsibility and the rule of law, which s. 51(e) of the Act is directed to.  Also in Gladue, at para. 78, this Court stated that it is unreasonable to assume that Aboriginal people do not believe in goals of punishment such as denunciation, deterrence and separation, to which I add, obviously, the principle of rehabilitation.  These goals of punishment, as discussed above, are related to the temporary disenfranchisement of serious criminal offenders and are ultimately aimed at the reintegration of offenders, Aboriginal or otherwise, back into the community.

205                           I also note that this Court stated in Gladue, at para. 79, that the more serious the offence, the more likely it will be that the terms of imprisonment for Aboriginal and non-Aboriginal persons will be similar or the same, even if the different considerations for sentencing are taken into account.  Since the temporary disenfranchisement provision hinges on serious criminal activity, this point is directly relevant.

206                           Having found that there is no infringement of s. 15(1)  of the Charter , it is unnecessary to consider s. 1.

VI.  Conclusion

207                           Section 51(e) of the Act does not infringe s. 15(1)  of the Charter .  While it has been conceded that it does infringe s. 3  of the Charter , the infringement is a reasonable limit that is demonstrably justified in a free and democratic society.  I would therefore dismiss the appeal.

208                           I would answer the constitutional questions as follows:

1.    Does s. 51(e) of the Canada Elections Act, R.S.C. 1985, c. E-2, infringe the right to vote in an election of members of the House of Commons, as guaranteed by s. 3  of the Canadian Charter of Rights and Freedoms ?

Yes; the infringement was conceded by the Crown.

2.    If the answer to Question 1 is yes, is the infringement a reasonable limit, prescribed by law, which can be demonstrably justified in a free and democratic society, pursuant to s. 1  of the Canadian Charter of Rights and Freedoms ?

Yes.

3.    Does s. 51(e) of the Canada Elections Act, R.S.C. 1985, c. E-2, infringe the right to equality before and under the law and equal benefit of the law without discrimination, as guaranteed by s. 15(1)  of the Canadian Charter of Rights and Freedoms ?

No.

4.    If the answer to Question 3 is yes, is the infringement a reasonable limit, prescribed by law, which can be demonstrably justified in a free and democratic society, pursuant to s. 1  of the Canadian Charter of Rights and Freedoms ?

It is not necessary to answer this question.

Appeal allowed, L’Heureux‑Dubé, Gonthier, Major and Bastarache JJ. dissenting.

Solicitor for the appellant Richard Sauvé:  Fergus J. O’Connor, Kingston, Ontario.

Solicitor for the appellants Sheldon McCorrister, Lloyd Knezacek, Clair Woodhouse, Aaron Spence, Serge Bélanger, Emile A. Bear and Randy Opoonechaw:  Public Interest Law Centre, Winnipeg.

Solicitor for the respondents:  The Department of Justice, Winnipeg.

Solicitors for the intervener the Attorney General of Alberta:  Fraser Milner Casgrain, Edmonton.

Solicitor for the intervener the Attorney General of Manitoba:  The Department of Justice, Winnipeg.

Solicitors for the interveners the Canadian Association of Elizabeth Fry Societies and the John Howard Society of Canada:  Allan Manson, Queen’s University, Kingston; Elizabeth Thomas, Kingston.

Solicitors for the intervener the British Columbia Civil Liberties Association:  Conroy & Co., Abbotsford, B.C.

Solicitor for the intervener the Aboriginal Legal Services of Toronto Inc.:  The Aboriginal Legal Services of Toronto — Legal Clinic, Toronto.

Solicitors for the intervener the Canadian Bar Association:  Desjardins Ducharme Stein Monast, Montréal.