Canada (Attorney General) v. Whaling, 2014 SCC 20, [2014] 1 SCR 392

  

 

SUPREME COURT OF CANADA

 

Citation: Canada (Attorney General) v. Whaling, 2014 SCC 20, [2014] 1 S.C.R. 392

Date: 20140320 Docket: 35024

 

Between:

Attorney General of Canada

Appellant

and

Christopher John Whaling

Respondent

And between:

Attorney General of Canada

Appellant

and

Judith Lynn Slobbe

Respondent

And between:

Attorney General of Canada

Appellant

and

Cesar Maidana

Respondent

- and -

Attorney General of Ontario and

British Columbia Civil Liberties Association

Interveners

 

 

Coram: McLachlin C.J. and LeBel, Abella, Rothstein, Cromwell, Moldaver, Karakatsanis and Wagner JJ.

 

Reasons for Judgment: (paras. 1 to 89)

Wagner J. (McLachlin C.J. and LeBel, Abella, Rothstein, Cromwell, Moldaver and Karakatsanis JJ. concurring)

 

 

 

Canada (Attorney General) v. Whaling, 2014 SCC 20, [2014] 1 S.C.R. 392

Attorney General of Canada                                                                         Appellant

v.

Christopher John Whaling                                                                         Respondent

‑ and ‑

Attorney General of Canada                                                                         Appellant

v.

Judith Lynn Slobbe                                                                                     Respondent

‑ and ‑

Attorney General of Canada                                                                         Appellant

v.

Cesar Maidana                                                                                            Respondent

and

Attorney General of Ontario and

British Columbia Civil Liberties Association                                             Interveners

Indexed as:  Canada (Attorney General) v. Whaling

2014 SCC 20

File No.:  35024.

2013:  October 15; 2014:  March 20.

Present:  McLachlin C.J. and LeBel, Abella, Rothstein, Cromwell, Moldaver, Karakatsanis and Wagner JJ.

on appeal from the court of appeal for british columbia

                    Constitutional law — Charter of Rights  — Double jeopardy — Repeal of early parole provisions applying retrospectively to offenders already sentenced — Whether retrospective application constitutes “punish[ment] . . . again” thereby infringing s. 11(h)  of Canadian Charter of Rights and Freedoms  — If so, whether infringement reasonable limit prescribed by law as can be demonstrably justified in free and democratic society under s. 1  of Charter  — Abolition of Early Parole Act, S.C. 2011, c. 11, s. 10(1).

                    W, S and M were all serving federal penitentiary sentences.  As first‑time, non‑violent offenders, all three were eligible for accelerated parole review (“APR”) under the system in place at the time of their sentencing.  With the coming into force of the Abolition of Early Parole Act (“AEPA”), APR was abolished.  Section 10(1) of the AEPA made the abolition of APR apply retrospectively to offenders already serving their sentences.  This changed the timing of eligibility for day parole: eligibility after the offender had served one sixth of the sentence or six months was replaced with eligibility six months before the full parole eligibility date.  Because the effect of the APR’s abolition was to delay the day parole eligibility dates of W, S and M, they challenged the constitutionality of s. 10(1). Both the trial judge and the Court of Appeal held that s. 10(1) infringed their right guaranteed by s. 11( h )  of the Charter  not to be “punished . . . again” for an offence and that the infringement was not saved under s. 1 .

                    Held:  The appeal should be dismissed.

                    Section 10(1) of the AEPA infringes s. 11( h )  of the Charter .  The introductory words to s. 11 provide that its subject is a “person charged with an offence”.  Paragraph (h) then provides that this person has the right, if found guilty and punished for the offence, not to be tried or punished for it again.  The disjunctive language of the words “tried or punished” indicates that s. 11(h)’s protection against additional punishment is independent of its protection against being tried again.  In other words, the protection applies to both the harassment of multiple trials and the harassment of additional punishment.  The conjunctive language of the words “found guilty and punished” further accentuates the disjunctive language of “tried or punished”.  It is thus clear from the plain meaning of the words that either being tried again or being punished again is sufficient to engage s. 11(h).

                    While the academic literature focuses on the fine points of what constitutes a second proceeding for the purposes of s. 11(h), this does not preclude its application to cases of “punish[ment] . . . again” in which no such proceeding took place.  If anything, the lack of literature on this subject speaks less to the scope of the provision than to the relative infrequency of such infringements.

                    Even in the few s. 11(h) cases, such as R. v. Wigglesworth, [1987] 2 S.C.R. 541, this Court found that the protection against double jeopardy could be triggered by proceedings that are criminal in nature or by “true penal consequences”. More recently, in R. v. Rodgers, 2006 SCC 15, [2006] 1 S.C.R. 554, the Court articulated a test for determining whether a given consequence or sanction constitutes punishment.  However, the question in this case eludes either test.  Rather than requiring us to determine whether a discrete sanction is punitive in nature, this case requires us to determine whether retrospective changes to parole eligibility, which modify the manner in which an existing sanction is carried out, constitute punishment.  The alleged punishment is neither a second proceeding nor a sanction in the sense contemplated in Rodgers.  Rather, the offender’s expectations about the original punishment or sanction have been frustrated and this is said to constitute new punishment.

                    The effect of every retrospective change will be context‑specific.  The dominant consideration in each case will be the extent to which an offender’s expectation of liberty has been thwarted by retrospective legislative action.  It is the retrospective frustration of an expectation of liberty that constitutes punishment.  Indeed, retrospective change that has the effect of automatically lengthening the offender’s period of incarceration represents one of the clearest of cases of retrospective double punishment under of s. 11(h).

                    Whether less drastic retrospective changes to parole constitute double punishment will depend on the circumstances of the particular case.  Generally speaking, a retrospective change to the conditions of a sentence will not be considered punitive if it does not substantially increase the risk of additional incarceration.  Indicators of a lower risk of additional incarceration include a process in which individualized decision making focused on the offender’s circumstances continues to prevail and procedural rights continue to be guaranteed in the determination of parole eligibility.  A change that directly results in an extension of the period of incarceration without regard to the offender’s individual circumstances and without procedural safeguards in the assessment process will clearly violate s. 11(h).

                    In this case, the purposes served by applying the AEPA to all offenders — rehabilitation, reintegration, public safety and confidence in the administration of justice — are not in issue.  However, the fact that Parliament had legitimate authority to legislate for such purposes does not shield the AEPA from constitutional scrutiny with regard to its effect.

                    The effect of the AEPA was to deprive W, S and M of the possibility of being considered for early day parole and to extend their minimum period of incarceration.  In this way, s. 10(1) had the effect of punishing W, S and M again. Because that effect was automatic and without regard to their individual circumstances, theirs is one of those “clearest of cases”.  Indeed, the imposition of delayed parole eligibility in this case is analogous to the imposition of delayed parole eligibility by a judge under the Criminal Code  as part of the sentence.  Imposing this same consequence by means of retrospective legislation triggers the protection against double punishment set out in s. 11(h).

                    The infringement by s. 10(1) of the AEPA of s. 11( h )  of the Charter  cannot be saved under s. 1 .  The AEPA’s objectives to reform parole administration and to maintain confidence in the justice system are pressing and substantial and its retrospective application is rationally connected to those objectives; however, the Crown has not discharged its burden of proving that there was no less intrusive an alternative to that retrospective application.  Indeed, prospective application — as opposed to retrospective — was an alternative available to Parliament that would have enabled it to attain its objectives without infringing s. 11(h).  The appeal is dismissed and the remedy ordered by the trial judge upheld.

Cases Cited

                    Distinguished:  R. v. Rodgers, 2006 SCC 15, [2006] 1 S.C.R. 554; R. v. Wigglesworth, [1987] 2 S.C.R. 541; referred to:  R. v. Shropshire, [1995] 4 S.C.R. 227; R. v. Chaisson, [1995] 2 S.C.R. 1118; R. v. Zinck, 2003 SCC 6, [2003] 1 S.C.R. 41; Cunningham v. Canada, [1993] 2 S.C.R. 143; Kienapple v. The Queen, [1975] 1 S.C.R. 729; R. v. Van Rassel, [1990] 1 S.C.R. 225; R. v. Shubley, [1990] 1 S.C.R. 3; R. v. M. (C.A.), [1996] 1 S.C.R. 500; R. v. Wust, 2000 SCC 18, [2000] 1 S.C.R. 455; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; R. v. Gamble, [1988] 2 S.C.R. 595; R. v. Harrer, [1995] 3 S.C.R. 562; R. v. Généreux, [1992] 1 S.C.R. 259; R. v. Pearson, [1992] 3 S.C.R. 665; Schachter v. Canada, [1992] 2 S.C.R. 679.

Statutes and Regulations Cited

Abolition of Early Parole Act, S.C. 2011, c. 11, ss. 3, 5, 10(1).

Canadian Charter of Rights and Freedoms , ss. 1 , 7 , 11( g ) , (h), (i).

Corrections and Conditional Release Act , S.C. 1992, c. 20, ss. 102 , 119 , 119.1  [ad. 1997, c. 17, s. 21(1); rep. 2011, c. 11, s. 3], 122, 125 [rep. 2011, c. 11, s. 5], 126 [idem], 126.1 [idem], 140.

Criminal Code , R.S.C. 1985, c. C‑46, s. 743.6(1) .

Treaties and Other International Instruments

International Covenant on Civil and Political Rights, 999 U.N.T.S. 171, art. 14.7.

Authors Cited

Canada.  House of Commons.  House of Commons Debates, vol. 145, No. 131, 3rd Sess., 40th Parl., February 15, 2011, p. 8205.

Canada.  House of Commons.  Standing Committee on Justice and Human Rights. Sub‑committee on Corrections and Conditional Release Act .  “A Work in Progress:  The Corrections and Conditional Release Act ”, May 2000 (online:  http://www.parl.gc.ca/HousePublications/Publication.aspx?DocId=2537364&Language=E&Mode=1&Parl=36&Ses=2).

Canadian Criminal Justice Association.  “Comments on ‘Directions for Reform’:  A Public Consultation Package on Sentencing, Corrections and Conditional Release”, December 7, 1990.

Correctional Service of Canada Review Panel.  Report of the Correctional Service of Canada Review Panel:  A Roadmap to Strengthening Public Safety.  Ottawa:  Minister of Public Works and Government Services Canada, 2007.

Friedland, M. L.  “Legal Rights Under The Charter” (1982), 24 Crim. L.Q. 430.

Hogg, Peter W.  Constitutional Law of Canada, 5th ed. Supp., vol. 2.  Toronto:  Carswell, 2007 (updated 2012, release 1).

Stuart, Don.  Charter Justice in Canadian Criminal Law, 5th ed.  Toronto:  Carswell, 2010.

                    APPEAL from a judgment of the British Columbia Court of Appeal (Levine, D. Smith and Groberman JJ.A.), 2012 BCCA 435, 329 B.C.A.C. 118, 560 W.A.C. 118, 98 C.R. (6th) 346, 292 C.C.C. (3d) 502, [2012] B.C.J. No. 2258 (QL), 2012 CarswellBC 3357, affirming a decision of Holmes J., 2012 BCSC 944, 264 C.R.R. (2d) 160, [2012] B.C.J. No. 1312 (QL), 2012 CarswellBC 1879.  Appeal dismissed.

                    Cheryl D. Mitchell and Ginette Gobeil, for the appellant.

                    Eric Purtzki and Garth Barriere, for the respondents.

                    David Lepofsky and Mabel Lai, for the intervener the Attorney General of Ontario.

                    Michael Jackson, Q.C., and Joana Thackeray, for the intervener the British Columbia Civil Liberties Association.

                    The judgment of the Court was delivered by

                    Wagner J. —

I.              Introduction

[1]                              In this appeal, the Court revisits the definition of the term “punishment” in the context of s. 11( h )  of the Canadian   Charter of Rights and Freedoms . The criminal law distinguishes between the sentence imposed on an offender and the conditions of the sentence. Changes to the conditions of a sentence, such as eligibility for parole, do not alter the sentence itself. This Court must decide whether retrospective changes to the conditions of a sentence may in some circumstances constitute “punishment” in violation of the s. 11(h) right not to be punished twice for the same offence.

[2]                              This appeal results from Parliament’s conclusion that accelerated parole review, or APR, was not working. Established by legislation enacted in November 1992, APR was a simplified process that allowed first-time non-violent offenders to be considered for parole on the basis of a single question: Are there no reasonable grounds to believe that the offender, if released, is likely to commit a violent offence? (See the Corrections and Conditional Release Act , S.C. 1992, c. 20  (“CCRA ”).)

[3]                              The Crown cites criticism of APR that dated back to its inception. Even before the legislation establishing APR was enacted, the Canadian Criminal Justice Association had expressed concerns about the violent recidivism test, noting that “[n]o system of predicting future behaviour can be safely based upon such a single factor.” It argued that APR would be available to some offenders who “have records with previous incarceration in Provincial institutions for offences including those of violence” (“Comments on ‘Directions for Reform’: A Public Consultation Package on Sentencing, Corrections and Conditional Release”, December 7, 1990, at p. 29 (A.R., vol. III, at p. 165)).

[4]                              Despite this and other criticism, the APR process was expanded in 1997 to include earlier eligibility for day parole: after six months, or one sixth of the sentence, whichever was longer, instead of six months before eligibility for full parole (S.C. 1997, c. 17, s. 21(1)).

[5]                              The Crown notes that criticism continued to be levelled at APR in the ensuing years. In a report presented in 2000, a parliamentary sub-committee, though recommending that APR be retained, also recommended “tightening the eligibility criteria” for APR so that offenders incarcerated for Schedule I or Schedule II offences under the CCRA  would not qualify for it, and “changing the risk of recidivism criterion to be taken into account by the National Parole Board” to one of general recidivism (Sub-committee on Corrections and Conditional Release Act  of the Standing Committee on Justice and Human Rights, “A Work in Progress: The Corrections and Conditional Release Act ”, May 2000, at para. 4.25 (online)).

[6]                              In a 2007 report, the Correctional Service of Canada Review Panel concluded that “statutory release and accelerated parole have both undermined discretionary release and generally have not proved as effective as discretionary release in mitigating violent reoffending” (Report of the Correctional Service of Canada Review Panel: A Roadmap to Strengthening Public Safety (2007), at p. 110).

[7]                              Purporting to respond to this criticism, Parliament repealed the relevant provisions. The Abolition of Early Parole Act, S.C. 2011, c. 11 (“AEPA”), the relevant part of which came into force on March 28, 2011, eliminated APR and with it the possibility of earlier release. What is crucial in this appeal is that by virtue of s. 10(1) of the AEPA, the repeal applied retrospectively.  Mr. Whaling, Ms. Slobbe and Mr. Maidana had all been convicted of serious, but non-violent, crimes at a time when APR remained in effect. Each of them would have been eligible for early day parole under the repealed provisions.

[8]                              The question before this Court is whether the retrospective application of the delayed eligibility for day parole to incarcerated offenders who had been sentenced before the APR provisions were repealed violated the respondents’ right, guaranteed by s. 11( h )  of the Charter , not to be punished anew for their offences.

[9]                              This appeal affords the Court the opportunity to revisit the purpose of s. 11(h) and to define its scope. For the reasons that follow, I find that s. 11(h) applies to the respondents’ claim. The retrospective application of delayed day parole eligibility violated the respondents’ s. 11(h) right not to be “punished . . . again”, and that violation was not justified under s. 1  of the Charter .

[10]                          Having found that the specific right enshrined in s. 11(h) was violated, I do not find it necessary to address the respondents’ claim with respect to the more general right guaranteed by s. 7  of the Charter .

II.           Facts

[11]                          This is an appeal from a decision in which the British Columbia Court of Appeal upheld a summary trial decision of the British Columbia Supreme Court. Both courts found in the respondents’ favour.

[12]                          The respondents, Mr. Whaling, Ms. Slobbe and Mr. Maidana, were all serving federal penitentiary sentences. As first-time, non-violent offenders, all three were eligible for APR under the system in place at the time of their sentencing (s. 125(1)  of the CCRA  (now repealed)).

[13]                          APR was different from normal parole review in a few ways. First, the process was simplified. The APR application was automatic, which meant that eligible offenders were referred to the National Parole Board without having to apply for it (s. 126(4)  of the CCRA  (repealed)). The review was conducted on paper, without a hearing (s. 126(1) (repealed)). Second, the test for release was based on a presumptive standard that was lower than the one applicable to normal parole. After finding “no reasonable grounds to believe that the offender, if released, is likely to commit an offence involving violence”, the Board had no discretion to decide against releasing the offender (s. 126(2) (repealed)). Section 126.1 (repealed) extended these provisions to day parole, in addition to full parole.

[14]                          Third, and this is crucial to this appeal, beginning in 1997, the APR process for day parole was triggered at an earlier date than in the normal process: after the offender had served one sixth of the sentence or six months (whichever was longer), instead of six months before the full parole eligibility date (s. 119.1 (repealed)).

[15]                          With the coming into force of the AEPA, APR (including early day parole eligibility), was abolished. Sections 3 and 5 of the AEPA repealed the CCRA ’s APR provisions (ss. 119.1, 125, 126 and 126.1, cited above), while s. 10(1) of the AEPA made the abolition of APR apply retrospectively to offenders already serving their sentences. Section 10(1) reads:

10. (1) Subject to subsection (2), the accelerated parole review process set out in sections 125  to 126.1  of the Corrections and Conditional Release Act , as those sections read on the day before the day on which section 5 comes into force, does not apply, as of that day, to offenders who were sentenced, committed or transferred to penitentiary, whether the sentencing, committal or transfer occurs before, on or after the day of that coming into force.

[16]                          In other words, APR, which the AEPA abolished, “does not apply” even if the offender was sentenced before the AEPA came into force. Instead of APR, the normal parole provisions of the CCRA  would now apply. This changed the timing of eligibility for day parole: eligibility after the offender had served one sixth of the sentence or six months (repealed s. 119.1) was replaced with eligibility six months before the full parole eligibility date (s. 119). It also changed the review process for both day and full parole: automatic referral to the Board (repealed ss. 126(4) and 126.1) was eliminated, which meant that the offender would have to submit an application (s. 122), and the review on paper without a hearing (repealed s. 126(1)) was replaced with the hearing and personal appearance by the offender required in the normal review process (s. 140). In addition, the repeal changed the test for granting parole: the lower, presumptive standard of violent recidivism, which left the Board no discretion to deny parole if the test was met (repealed s. 126(2)), reverted to the more onerous one of “undue risk to society”, which does leave it with such a discretion (s. 102).

[17]                          The immediate effect of the repeal was to delay the day parole eligibility dates of all three respondents: Mr. Whaling’s by three months, Ms. Slobbe’s by nine months, and Mr. Maidana’s by twenty-one months.

[18]                          The respondents challenged the constitutionality of s. 10(1) of the AEPA by way of a summary trial in the British Columbia Supreme Court.

III.        Judicial History

A.           British Columbia Supreme Court, 2012 BCSC 944, 264 C.R.R. (2d) 160

[19]                          Holmes J., the summary trial judge, held that s. 10(1) of the AEPA infringed s. 11( h )  of the Charter  because it amounted to additional punishment, and that it was not saved under s. 1  of the Charter .

[20]                          Holmes J. began by asking whether the abolition of APR amounted to punishment in violation of the respondents’ right under s. 11( h )  of the Charter , “if finally found guilty and punished for [an] offence, not to be tried or punished for it again”. She reasoned on the basis of R. v. Rodgers, 2006 SCC 15, [2006] 1 S.C.R. 554, that “[s.] 11(h)’s protections carry on past trial, conviction (or acquittal), and sentence, to prevent any additional trial or punishment for the offence with which the person was once charged” (para. 46).

[21]                          On how to determine whether a given consequence constitutes “punishment”, Holmes J. quoted (at para. 52) the following passage from Rodgers (para. 63):

As a general rule, it seems to me that the consequence will constitute a punishment when it forms part of the arsenal of sanctions to which an accused may be liable in respect of a particular offence and the sanction is one imposed in furtherance of the purpose and principles of sentencing.

She pointed out that delayed parole eligibility is clearly used as “punishment” in the context of criminal sentencing, citing several decisions of this Court, including R. v. Shropshire, [1995] 4 S.C.R. 227, R. v. Chaisson, [1995] 2 S.C.R. 1118, and R. v. Zinck, 2003 SCC 6, [2003] 1 S.C.R. 41. Although there is a difference between the sentencing and parole processes, parole ineligibility imposed as part of a sentence under the Criminal Code , R.S.C. 1985, c. C-46 , is not materially different from parole ineligibility resulting from the retrospective amendment of the CCRA , since the objectives and functions of the two processes can and do overlap.

[22]                          Turning to s. 10(1) of the AEPA, Holmes J. found that there was “abundant evidence” that its purposes were punitive in nature (para. 112). She also found that the effect of the retrospective alteration of day parole eligibility was punitive, and noted in particular that this alteration would have a “significant actual effect on the way in which the two offenders who are still in prison will serve their sentences” (para. 113). Although she accepted, citing Cunningham v. Canada, [1993] 2 S.C.R. 143, that there are changes that can be made to corrections and parole law and policy without distorting the sentence imposed, in her view, “such changes do not include significant limitations, regardless of any exercise of the Board’s discretion, to the parole eligibility of offenders” who have already been sentenced (para. 114). A change that includes such limitations occasions additional punishment.

[23]                          Having established an infringement of s. 11(h), Holmes J. found that the infringement was not saved under s. 1  of the Charter . Although the objectives underlying the retrospective application of the repeal, which included maintaining or restoring public confidence in the administration of justice, were pressing and substantial, the Attorney General of Canada had not shown that the provision impaired the respondents’ s. 11(h) rights no more than was necessary to attain Parliament’s objectives. Holmes J. declared s. 10(1) to be invalid to the extent that it made the AEPA apply retrospectively to offenders sentenced before March 28, 2011, the date the AEPA came into force. The declaration was to have immediate effect.

B.            British Columbia Court of Appeal, 2012 BCCA 435, 329 B.C.A.C. 118 (Levine, D. Smith and Groberman JJ.A.)

[24]                          The Court of Appeal upheld the trial judge’s decision, although with a few variations in reasoning. Levine J.A., writing for the court, found that the purpose of s. 11(h) is to protect against double jeopardy, and that it applies to sanctions imposed after sentencing. She found that the retrospective legislative lengthening of the time the respondents would spend in jail that resulted from the delaying of their parole eligibility violated their right under s. 11(h) not to be “punished again” (para. 71).

[25]                          Levine J.A. emphasized the distinction between the role of the sentencing judge and that of the Board under the CCRA , and mentioned that it is improper for a judge to take parole administration into consideration in the sentencing process. This did not alter her conclusion that s. 10(1) of the AEPA nonetheless amounted to punishment, having regard to both the purpose and the effect of the retrospective application of the repeal.

[26]                          Levine J.A. noted that an overly narrow reading of s. 11(h) is contrary to the appropriate principles of Charter  interpretation. She stated that the purpose of s. 11(h) is to ensure that a person cannot be punished a second time for an offence for which he or she has already been tried and convicted. The issue was the meaning of “punishment” for the purposes of s. 11(h), and this turned on whether the consequence in this case “forms part of the arsenal of [possible] sanctions” and furthers the “purpose and principles of sentencing”, as required by Rodgers (para. 63). Levine J.A. asked whether the statute had a punitive purpose, but declined to decide the s. 11(h) claim on the basis of this question. Instead, she held that the effect of the delayed parole eligibility imposed by way of legislation was analogous to that of delayed parole eligibility imposed by a judge, the latter of which is clearly punishment (citing Chaisson and Zinck). In its effect, therefore, s. 10(1) of the AEPA violated s. 11( h )  of the Charter .

[27]                          The Court of Appeal also agreed with the trial judge’s finding that although the legislation was enacted in response to a pressing and substantial concern, the Attorney General had not provided sufficient evidence that it impaired the respondents’ rights as little as possible, which meant that it could not be saved under s. 1  of the Charter .

IV.        Analysis

A.           Does the Retrospective Application of the Abolition of APR Infringe Section 11(h)?

[28]                          The respondents submit that the retrospective application provision, s. 10(1) of the AEPA, violates their right under s. 11( h )  of the Charter  not to be “punished . . . again”.  They urge a broad reading of the term “punishment” that is not limited to the duplication of criminal or quasi-criminal proceedings. In their view, the retrospective application of the repeal, which eliminated the eligibility for early day parole of offenders who had already been sentenced, was punitive in its effect. The respondents also argue, parting ways with the Court of Appeal on this point, that it was punitive in its purpose.

[29]                          The Crown urges a narrow, textual reading of s. 11(h) that would exclude the elimination of early day parole eligibility from the definition of “punishment”. It argues that the retrospective application of the repeal was adopted in furtherance of the goals of rehabilitation, reintegration, public safety and confidence in the administration of justice, not that of punishment. The effect of that application reflects this non-punitive purpose. In this Court, the Crown further argues that s. 11(h) is not engaged, because being “punished . . . again” requires a duplication of proceedings that are criminal in nature in respect of the same matter.

[30]                          I will begin with this last question of whether s. 11(h) applies absent a duplication of proceedings. I will then consider the scope of the “punishment” concept in the context of s. 11(h) before inquiring into whether the impugned provision constitutes “punishment” in its purpose or in its effect. 

(1)         Does Section 11(h) Apply Where No Duplication of Proceedings Has Occurred?

[31]                          Neither the trial court nor the Court of Appeal dealt with this question. The two courts agreed that s. 11(h) protects against double jeopardy, which Levine J.A. defined as protection “from being tried or punished again for an offence for which the offender has already been found guilty and punished” (para. 45), and they focused their analyses on the meaning of “punishment” for the purposes of s. 11(h).

[32]                          Section 11( h )  of the Charter  reads as follows:

11. Any person charged with an offence has the right

 

. . .

 

(h) if finally acquitted of the offence, not to be tried for it again and, if finally found guilty and punished for the offence, not to be tried or punished for it again;

[33]                          Parliament’s purpose in enacting s. 11(h) was to protect against double jeopardy. Section 11(h) mirrors the language and purpose of art. 14.7 of the International Covenant on Civil and Political Rights, 999 U.N.T.S. 171, which reads:

7. No one shall be liable to be tried or punished again for an offence for which he has already been finally convicted or acquitted in accordance with the law and penal procedure of each country.

[34]                          But equating s. 11(h) with double jeopardy does not conclude the discussion of its purpose, since the very definition of “double jeopardy” is contested. Don Stuart says the following in this regard in Charter Justice in Canadian Criminal Law (5th ed. 2010):

Under existing Canadian law there is certainly no one rule on double jeopardy. The subject is one of the utmost complexity and subtlety and is certainly in need of clarification. The law provides protection against harassment of multiple trials for the same act but also protection against multiple punishment. The concern to do something about double punishment stems from a distinct consideration based far more squarely on the fairness of proportionate punishment. [p. 464]

[35]                          As several authors have noted, the scope of s. 11(h) is narrow (see M. L. Friedland, “Legal Rights Under The Charter” (1982), 24 Crim. L.Q. 430, at pp. 435 and 449; Stuart, at p. 467). Stuart states that s. 11(h) has had “little impact on the protection of the accused against double jeopardy and double punishment”, in part because of this narrow scope (ibid.). Both Friedland and Stuart suggest that broader interpretations of double jeopardy may fit more easily into s. 7  of the Charter  (Friedland, at p. 435; Stuart, at p. 468). This view may seem especially compelling in this case, which concerns alleged punishment arising from retrospective legislation, given that there are two provisions that deal explicitly with retrospectivity: s. 11(g), which protects against retroactive criminal legislation, and s. 11(i), which protects against the imposition of a harsher punishment where the punishment for the offence has been varied between the time of commission of the crime and the time of sentencing.

[36]                          In my view, it is not necessary to resort to a different Charter  provision. The language of s. 11(h), the academic literature and this Court’s jurisprudence support a reading of s. 11(h) according to which the right not to be “punished . . . again” applies where an offender has been sentenced, even if no separate proceeding has taken place.

[37]                          Let me begin by addressing the plain meaning of s. 11(h). The introductory words to s. 11 indicate that the subject of the entire section is a “person charged with an offence”. Paragraph (h) then provides that this person has the right, “if finally found guilty and punished for the offence, not to be tried or punished for it again”.  The disjunctive language of the words “tried or punished” clearly indicates that s. 11(h)’s protection against additional punishment is independent of its protection against being tried again. In other words, as Stuart notes in respect of double jeopardy more generally, the protection applies to both the harassment of multiple trials and the harassment of additional punishment (p. 464). The conjunctive language of the words “found guilty and punished” further accentuates the disjunctive language of “tried or punished”. It is thus clear from the plain meaning of the words that either being tried again or being punished again is sufficient to engage s. 11(h).

[38]                          The plain meaning of s. 11(h) is supported by common sense. It would be far more questionable to punish someone without a proceeding than to punish him or her with a proceeding. The purpose of s. 11(h) cannot be to protect against punishment imposed following a trial in which due process has been observed, but not against punishment imposed without the protections afforded by a trial. 

[39]                          Next, it is true that the authors emphasize that s. 11(h) is narrow in scope and does not encompass all forms of double jeopardy. For instance, it is distinct from statutory and common law protections such as autrefois acquit, or the rule in Kienapple v. The Queen, [1975] 1 S.C.R. 729, against multiple convictions for separate offences based on a single act (R. v. Van Rassel, [1990] 1 S.C.R. 225, at p. 233). Nevertheless, this does not preclude the application of s. 11(h) to protect against double punishment where no new proceeding has taken place, nor am I aware of academic literature that would support such a position. While some authors focus on the fine points of what constitutes a second proceeding for the purposes of s. 11(h) (see P. W. Hogg, Constitutional Law of Canada (5th ed. Supp.), vol. 2, at p. 51-35), this does not limit the application of s. 11(h) in cases of double punishment in which no such proceeding took place.

[40]                          If anything, the lack of academic commentary on this subject speaks less to the scope of the provision than to the relative infrequency of such violations. The early cases with respect to s. 11(h) concerned situations in which criminal charges were laid against an accused who had previously been sanctioned or disciplined in a non-criminal proceeding for the same act; the question was whether the non-criminal proceeding or sanction triggered the protection against double jeopardy (R. v. Wigglesworth, [1987] 2 S.C.R. 541, and R. v. Shubley, [1990] 1 S.C.R. 3). Those cases did not directly address the possibility of punishment without a second proceeding. The Crown cites the following comment made by McLachlin J. (as she then was) in Shubley (p. 19): “Section 11(h) provides protection against duplication in proceedings of a criminal nature. It does not preclude two different proceedings, one criminal and the other not criminal, flowing from the same act” (A.F., at para. 59). However, this passage must be read in context, as part of the Court’s inquiry into whether the proceeding in question was criminal “in nature”. Even in the early s. 11(h) cases, it was found that the protection against double jeopardy could be triggered by proceedings that are criminal in nature or by “true penal consequences”. 

[41]                          Rodgers, a more recent case, supports my conclusion. At issue in that case was whether an order for the taking of a DNA sample that was made pursuant to a statute enacted after the offender had been sentenced constituted double punishment. The following comments of Charron J. are on point:

First, it is necessary to consider whether s. 11 applies at all to a s. 487.055 application.  As the introductory words of the section make it plain, the protection extended by s. 11 can only be invoked when “[a] person [is] charged with an offence”. Therefore, in and of itself, the application for a DNA order does not at all engage s. 11. It cannot be contended that Mr. Rodgers is “charged with an offence” on any reasonable meaning of the term and, as I understood his argument, he is not claiming the protection of s. 11 on that basis.  He relies, rather, on the charges that were brought in respect of the index offences — namely the multiple sex offences in respect of which he was convicted and which form the basis of the application for a DNA data bank order.  There is no doubt that s. 11 applies to those criminal proceedings and the question then becomes whether the imposition of a s. 487.055 order constitutes further “punishment” for those offences. [Emphasis added; para. 58.]

In Rodgers, the Court rejected the narrow construction of s. 11(h) advanced by the Crown in the case at bar according to which it is limited to cases involving a duplication of proceedings that are criminal in nature. I find that s. 11(h) does protect an offender who has been tried and sentenced against double punishment even in the absence of a second proceeding.  As to the paucity of academic commentary on this issue, it may be time to update the textbooks.

[42]                          Having concluded that s. 11(h) does not preclude claims of double punishment where a second proceeding has not taken place, I must now determine whether the situation in the instant case amounts to punishment.

(2)         What Is the Scope of “Punishment” in the Context of Section 11(h)?

(a)          Pre-Rodgers Jurisprudence

[43]                          The scope of “punishment” in the context of s. 11(h) has expanded over the years as new cases have pushed the limits of old definitions. It has always been clear that criminal or quasi-criminal proceedings trigger the protection against double jeopardy. Thus, a second criminal or quasi-criminal charge with respect to the same act engages s. 11(h) even if the consequence is slight.

[44]                          Wigglesworth made it clear that the protection against double jeopardy may be triggered not only by proceedings that are criminal or quasi-criminal in nature, but also by non-criminal proceedings that result in a sanction with true penal consequences. Where a person is charged in respect of “a private, domestic or disciplinary matter . . . intended to maintain discipline, integrity or to regulate conduct within a limited private sphere of activity” (p. 560), s. 11(h) may still be engaged if the true penal consequences test is met:

In my opinion, a true penal consequence which would attract the application of s. 11 is imprisonment or a fine which by its magnitude would appear to be imposed for the purpose of redressing the wrong done to society at large rather than to the maintenance of internal discipline within the limited sphere of activity.

 

(Wigglesworth, at p. 561, per Wilson J.)

[45]                          In Wigglesworth, although the internal disciplinary proceeding against an RCMP officer was neither criminal nor quasi-criminal “in nature”, it was found to involve “true penal consequences” because the possible sanctions included imprisonment for up to one year.  In Shubley, applying the same test, the Court found that a sanction imposed in an internal prison disciplinary proceeding — close confinement for a period of five days on a special diet — fell short of “true penal consequences”, in part because the possible sanctions involved neither fines nor imprisonment: “Confined as they are to the manner in which the inmate serves his time, and involving neither punitive fines nor a sentence of imprisonment, they appear to be entirely commensurate with the goal of fostering internal prison discipline and are not of a magnitude or consequence that would be expected for redressing wrongs done to society at large” (p. 23, per McLachlin J.).

[46]                          In Rodgers, the Court took the opportunity to revisit the definition of “punishment” it had articulated in Wigglesworth and Shubley. The question was whether compelling an offender to submit to the taking of a DNA sample under legislation that had not existed at the time of his or her conviction constituted double punishment. Charron J. accepted that imprisonment and heavy fines constitute “true penal consequences”, but reasoned that “punishment” is not limited to these two types of sanctions. She set out to identify features characteristic of punitive sanctions that could be used to determine whether a given sanction that is less severe than imprisonment or a heavy fine nevertheless constitutes punishment. She mentioned various orders a sentencing court might make, such as an order for forfeiture, a firearm prohibition, a driving prohibition or an order for restitution.

[47]                          The two-part definition of punishment articulated by Charron J. in Rodgers, and relied upon in the instant case by the courts below and by the parties, is the following:

As a general rule, it seems to me that the consequence will constitute a punishment when it forms part of the arsenal of sanctions to which an accused may be liable in respect of a particular offence and the sanction is one imposed in furtherance of the purpose and principles of sentencing. [para. 63]

Turning to the case before her, Charron J. found that a DNA order is no more a part of the arsenal of criminal sanctions than the taking of a photograph or fingerprints. It does not on its own attach a stigma to the offender, and while it may have a deterrent effect, not all deterrent measures can be characterized as punishment. By way of example, Charron J. cites random traffic stops to check for alcohol consumption, which hopefully deter people from drinking and driving but are not punishment. 

(b)         Inapplicability of the Rodgers Test to Retrospective Changes to Parole Eligibility

[48]                          In my view, the case at bar once again pushes the limits of the “punishment” concept, requiring us to revisit the principles that define the scope of s. 11(h). Whereas in Wigglesworth the Court established that non-criminal proceedings may engage s. 11(h) if they result in true penal consequences, and in Rodgers it articulated a test for determining whether a given consequence or sanction constitutes punishment, the question in the instant case eludes both tests.

[49]                          Rather than requiring us to determine whether a discrete sanction is punitive in nature, this case requires us to determine whether retrospective changes to parole eligibility, which modify the manner in which an existing sanction is carried out, constitute punishment. The alleged punishment is neither a second proceeding nor a “sanction” in the sense contemplated in Rodgers. Rather, the offender’s expectations about the original punishment or sanction have been frustrated and this is said to constitute new punishment.

[50]                          The following makes clear that the two-part Rodgers test cannot apply to determine whether retrospective changes to parole eligibility constitute punishment. In the first branch of the Rodgers test, “punishment” is defined by referring to the traditional forms of punishment provided for in the Criminal Code , in which “[t]he words ‘sentence’ and ‘sanction’ are . . . used interchangeably” (para. 62). The function of this branch of the test is to enable courts to determine whether other types of sanctions — such as a DNA order or a driving prohibition — share the characteristics of punitive sanctions and thus constitute “punishment”. Since “a grant of parole represents a change in the conditions under which a judicial sentence must be served, rather than a reduction of the judicial sentence itself” (R. v. M. (C.A.), [1996] 1 S.C.R. 500, at para. 62 (emphasis deleted)), changes to the parole system do not generally form part of the “arsenal of sanctions” contemplated in Rodgers.

[51]                          On the other hand, a retrospective change to parole eligibility may have the effect of extending an offender’s term of incarceration. Incarceration is “the most severe deprivation of liberty known to our law” (Wigglesworth, at p. 562), and the most obvious example of punishment in the “arsenal of sanctions” available under the Criminal Code . Incarceration and heavy fines are the benchmark sanctions against which other, less severe sanctions are assessed under the Rodgers test. That incarceration constitutes “punishment” is a core underlying assumption of the Rodgers test.

[52]                          In short, when applied in this context, the Rodgers test is overly formalistic, as the “arsenal of sanctions” test would exclude most changes to parole eligibility, whereas even marginal increases in the likelihood of additional incarceration easily meet the test. The problem is that the Rodgers test does not assist in identifying situations in which, from a functional rather than a formalistic perspective, the harshness of punishment has been increased. The Rodgers test was designed for a different purpose, namely to determine whether a discrete sanction — one that does not modify the original sanction — has the characteristics of a criminal sanction, and thus constitutes “punishment”.

[53]                          The second branch of the Rodgers test — whether the sanction furthers the purpose and principles of sentencing — is also an awkward fit in this situation. For instance, rehabilitation and reintegration are principles applied in parole administration, but they are also sentencing principles. It would seem that all retrospective changes to parole eligibility could therefore be said to further the purpose and principles of sentencing, though it is clear to me that not all such changes would be punitive.

(c)          Retrospective Punishment and Double Jeopardy Under Section 11(h)

[54]                          In the cases leading up to Rodgers and in Rodgers itself, the Court identified two types of situations in which the rule against double jeopardy set out in s. 11(h) may be violated, but the case at bar introduces a third. In my view, where an offender has been finally acquitted of, or finally found guilty and punished for, an offence, s. 11(h) precludes the following further state actions in relation to the same offence:

(a)                a proceeding that is criminal or quasi-criminal in nature (being “tried . . . again”);

(b)               an additional sanction or consequence that meets the two-part Rodgers test for punishment (being “punished . . . again”) in that it is similar in nature to the types of sanctions available under the Criminal Code  and is imposed in furtherance of the purpose and principles of sentencing; and

(c)                retrospective changes to the conditions of the original sanction which have the effect of adding to the offender’s punishment (being “punished . . . again”).

The case at bar concerns the third of these types of double punishment under s. 11(h). It is not the repeal of the APR provisions that is alleged to be unconstitutional, but the retrospective application of that repeal, which altered the parole expectations of offenders who had already been sentenced.

[55]                          Before discussing the scope of double punishment in this context, I should comment briefly on the relationship between retrospective punishment and double jeopardy in the Charter  context. In addition to the common law presumption against retroactivity, there are principles of non-retroactivity specific to the criminal law that are enshrined in the Charter . The clearest provision in this regard is s. 11(g), which protects against retroactive criminal laws. As Peter Hogg writes, “[a]part from s. 11(g), Canadian constitutional law contains no prohibition of retroactive (or ex post facto) laws” (p. 51-33). Section 11(i), although less explicitly concerned with retroactivity, protects against increases in punishment between the time of commission of the crime and the time of sentencing. Both these provisions express society’s repudiation of retroactive punishment, broadly defined — of retroactive legislation establishing a criminal offence in the case of s. 11(g), and of retroactive legislation under which a harsher penalty would apply to an offence committed before its enactment in the case of s. 11(i).

[56]                          Section 11(h) is not expressly concerned with retroactivity. Its purpose is to protect against double jeopardy, namely the trial or punishment of an accused for an offence he or she has already been acquitted of or found guilty and punished for. In many, if not most, situations in which s. 11(h) applies, retroactive punishment will not be in issue. For instance, the issue in the early cases on s. 11(h) was whether a person who had already been sanctioned in a non-criminal proceeding could subsequently be charged for the same act under the Criminal Code . The offences in question existed at the time the acts were committed. However, as I explained above, s. 11(h) may be engaged where no duplication of proceedings has occurred. Retrospective modification of the parole system after an offender was sentenced may have the effect of increasing the offender’s punishment, thereby engaging s. 11(h). A legislative change that is not in itself punitive can acquire a punitive character by being applied retrospectively.

(d)         What Retrospective Changes to the Conditions of a Sentence Constitute Double Punishment?

[57]                          Generally speaking, offenders have constitutionally protected expectations as to the duration, but not the conditions, of their sentences. Various changes in the management of an offender’s parole are not punitive, even though they may engage the offender’s liberty interest by marginally increasing the likelihood of additional incarceration. McLachlin J. (as she then was) held as follows in Cunningham:

The Charter does not protect against insignificant or “trivial” limitations of rights . . . .  It follows that qualification of a prisoner’s expectation of liberty does not necessarily bring the matter within the purview of s. 7  of the Charter .  The qualification must be significant enough to warrant constitutional protection.  To require that all changes to the manner in which a sentence is served be in accordance with the principles of fundamental justice would trivialize the protections under the Charter .  To quote Lamer J. in Dumas [v. Leclerc Institute, [1986] 2 S.C.R. 459], at p. 464, there must be a “substantial change in conditions amounting to a further deprivation of liberty”. [p. 151]

[58]                          The requirement of a “substantial change in conditions amounting to a further deprivation of liberty” was articulated in the context of s. 7, and I will not import it into that of s. 11(h), the purpose of which is distinct. Retrospective changes to the parole system that engage a liberty interest under s. 7 will not necessarily constitute punishment for the purposes of s. 11(h). However, certain of the conclusions reached in Cunningham do apply to my analysis under s. 11(h). First of all, the Court recognized that an offender has an expectation of liberty that is based on the parole system in place at the time of his or her sentencing, and that thwarting that expectation may engage a constitutionally protected liberty interest. Changes to the parole system that add retrospectively to the offender’s incarceration may violate s. 7 even if they do not affect the sentence itself. As McLachlin J. put it in Cunningham: “One has ‘more’ liberty, or a better quality of liberty, when one is serving time on mandatory supervision than when one is serving time in prison” (p. 150).

[59]                          This being said, the Court recognized in Cunningham that not all expectations of liberty in the parole context are constitutionally protected. Even where a change to the conditions of a sentence engages a liberty interest under s. 7, it may nonetheless be consistent with the principles of fundamental justice. Some line drawing becomes necessary. In my view, this same basic point applies to a retrospective change that constitutes double punishment in the s. 11(h) context. Some retrospective changes to the parole system affect the expectation of liberty of an offender who has already been sentenced to such an extent that they amount to new punishment, while other changes have a more limited impact and do not trigger Charter  protection.

[60]                          I will not articulate a formula that would apply to every case, because such a formula is not needed to resolve this appeal and the effect of every retrospective change will be context-specific. That said, the dominant consideration in each case will in my view be the extent to which an offender’s settled expectation of liberty has been thwarted by retrospective legislative action. It is the retrospective frustration of an expectation of liberty that constitutes punishment. At one extreme, a retrospective change to the rules governing parole eligibility that has the effect of automatically lengthening the offender’s period of incarceration constitutes additional punishment contrary to s. 11( h )  of the Charter . A change that so categorically thwarts the expectation of liberty of an offender who has already been sentenced qualifies as one of the clearest of cases of a retrospective change that constitutes double punishment in the context of s. 11(h).

[61]                          I reach this conclusion on the basis of many of the reasons cited by the courts below and advanced by the respondents. Although a sentencing judge is not to consider parole eligibility in assessing the fitness of the sentence (Zinck, at para. 18), the punitive effect of delayed parole eligibility is expressly recognized in the Criminal Code , which empowers a sentencing judge to consider delayed parole eligibility to be part of the sentence in certain circumstances. For example, s. 743.6(1) empowers a court to impose delayed parole eligibility for the purpose of denunciation or of specific or general deterrence. Furthermore, a sentencing judge may increase the parole ineligibility period of an offender convicted of second degree murder (Chaisson, at para. 12). The Court said the following in this regard in R. v. Wust, 2000 SCC 18, [2000] 1 S.C.R. 455, at para. 24:

Rarely is the sentencing court concerned with what happens after the sentence is imposed, that is, in the administration of the sentence. Sometimes it is required to do so by addressing, by way of recommendation, or in mandatory terms, a particular form of treatment for the offender. For instance in murder cases, the sentencing court will determine a fixed term of parole ineligibility: s. 745.4 of the Code. [Emphasis added.]

In Shropshire, Iacobucci J. noted that the duration of parole ineligibility is the only difference in terms of punishment between first and second degree murder, which “clearly indicates that parole ineligibility is part of the ‘punishment’ and thereby forms an important element of sentencing policy” (para. 23).

[62]                          The fact that delayed parole eligibility can be imposed in the sentencing process confirms my view that retrospectively imposing delayed parole eligibility on offenders who have already been sentenced constitutes punishment. Where Parliament imposes through retrospective legislation a consequence that sentencing judges may themselves impose for the purpose of punishment, the s. 11(h) protection against double punishment applies.

[63]                          Whether less drastic retrospective changes to parole constitute double punishment will depend on the circumstances of the particular case. Generally speaking, a retrospective change to the conditions of a sentence will not be considered punitive if it does not substantially increase the risk of additional incarceration. Indicators of a lower risk of additional incarceration include a process in which individualized decision making focused on the offender’s circumstances continues to prevail and procedural rights continue to be guaranteed in the determination of parole eligibility. Though I caution against directly importing principles drawn from the s. 7 jurisprudence into this context, the replacement of an automatic release system with a discretionary release system was found to be constitutional in Cunningham owing in part to various procedural safeguards, including a hearing and the entitlement to counsel. While s. 11(h) is not directly concerned with procedural safeguards, the presence or absence of such safeguards is relevant in considering the likelihood of the punishment’s severity being increased. As I mentioned above, the dominant consideration will be the extent to which the offender’s settled expectation of liberty has been thwarted. A change that directly results in an extension of the period of incarceration without regard to the offender’s individual circumstances and without procedural safeguards in the assessment process will clearly violate s. 11(h).

(3)         Does Section 10(1) of the AEPA Violate Section 11(h) of the Charter  in Its Purpose or in Its Effect?

[64]                          A law may violate the Charter  in its purpose or in its effect: “If the legislation fails the purpose test, there is no need to consider further its effects, since it has already been demonstrated to be invalid” (R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 334). In determining whether s. 10(1) of the AEPA is compatible with s. 11( h )  of the Charter , both the trial judge and the Court of Appeal considered the purpose of the law first, followed by its effect. The trial judge held that s. 10(1) of the AEPA violates the Charter  in both its purpose and its effect, whereas the Court of Appeal declined, because the effect of s. 10(1) renders it unconstitutional, to rule on the constitutionality of its purpose. I agree with the Court of Appeal, but for different reasons, that s. 10(1) violates the Charter  in its effect.

(a)          Purpose of Section 10(1) of the AEPA

[65]                          First of all, the overall purpose of the AEPA is not at issue. I agree with the Court of Appeal, which found that there is no serious dispute about the validity of that purpose:

The Attorney General points to evidence it provided at trial that supports repealing APR to enhance the purposes and principles of parole regimes: rehabilitating and successfully reintegrating offenders by shifting away from presumptive release to normal parole; maintaining public safety and reducing recidivism; shifting the focus of parole review from offence-based presumptive release to parole consideration based on the circumstances of the offender; and maintaining or restoring confidence in the administration of justice by ensuring sentences as administered reflect the sentences courts impose.  These purposes are entirely within the constitutional powers of the government, and the respondents take no issue with the AEPA on that basis. [para. 54]

In other words, the purposes served by limiting the parole eligibility of all offenders are within the prerogative of Parliament and, in any event, are not at issue in this case. What is at issue is Parliament’s decision to make the repeal apply retrospectively, by means of s. 10(1) of the AEPA, to offenders who had already been sentenced. What is the purpose of this retrospective application? 

[66]                          The foregoing discussion leads me to conclude that where legislation makes retrospective changes to parole eligibility, it will violate s. 11(h) if its purpose is to prolong the offender’s period of incarceration. The Crown submits that Parliament’s decision to have the repeal apply retrospectively was made in furtherance of the goals of rehabilitation, reintegration, public safety and confidence in the administration of justice. These goals were better served by having the repeal apply “immediately and uniformly to all offenders, including those already sentenced but not yet released” (A.F., at para. 84).  The respondents contend, and the trial judge so held, that this argument conceals the true purpose of the retrospective application provision. Both they and Holmes J. have cited Hansard to show that retrospective application was meant to serve the purposes of denunciation, deterrence and punishment. These arguments circle around the central question: Was Parliament’s purpose to lengthen the time to be served in jail by offenders who had already been sentenced?

[67]                          In particular, one Member of Parliament had said that if the repeal was not retroactive, the victims of a notorious crime committed before the repeal “will never have any kind of justice served” (House of Commons Debates, vol. 145, No. 131, 3rd Sess., 40th Parl., February 15, 2011, at p. 8205). This strongly suggests that Parliament intended to extend the period of incarceration of offenders who had already been sentenced, which would mean that the purpose of the provision was incompatible with s. 11(h).

[68]                          Despite some troubling passages from Hansard that are suggestive of an unconstitutional purpose, I accept the Crown’s articulation of Parliament’s purpose for the following reasons. Parole administration is distinct from sentencing. Parliament, responding to criticism of APR, saw fit to adopt a different approach by returning to the normal pre-APR parole rules, and considered it desirable to pursue the objectives underlying those rules in respect of all offenders, including those who had already been sentenced. In light of the formal distinction between sentencing and parole administration, I see nothing prima facie unconstitutional about the purpose of applying legitimate parole objectives universally to all offenders.

[69]                          However, the fact that Parliament had legitimate authority to legislate in relation to parole within the framework of the CCRA  does not shield the legislation in question from constitutional scrutiny with regard to its effect. Parliament’s authority to modify the conditions of an existing sentence, which is rooted in the formal distinction between parole administration and sentencing, is limited where the effect of a modification is such that it constitutes punishment.

(b)         Effect of Section 10(1) of the AEPA

[70]                          The effect of the retrospective application provision, s. 10(1) of the AEPA, was to deprive the three respondents of the possibility of being considered for early day parole, which was an expectation they had had at the time they were sentenced. This amounts to a lengthening of the minimum period of incarceration for persons — like the respondents — who would have qualified for early day parole under the APR system.

[71]                          In my view, s. 10(1) had the effect of punishing the respondents again. It retrospectively imposed a delay in day parole eligibility in relation to offences for which they had already been tried and punished. The effect — extended incarceration — was automatic and without regard to individual circumstances.

[72]                          This situation is one of the “clearest of cases” discussed above. The imposition of a delay in parole eligibility in this case is analogous to the imposition of delayed parole eligibility by a judge under the Criminal Code  as part of the sentence. As I mentioned above, Iacobucci J. noted in Shropshire that the imposition of such a delay “clearly indicates that parole ineligibility is part of the ‘punishment’” (para. 23). Imposing this same consequence by means of retrospective legislation triggers the protection against double punishment set out in s. 11(h).

[73]                          My conclusion is not altered by the fact that the extension of parole ineligibility was imposed by means of legislation in the context of the CCRA , which concerns, inter alia, sentence administration, as opposed to being judicially imposed under the Criminal Code . This formal distinction does not change the basic point that, from a functional perspective, the new period of parole ineligibility thwarted the expectations of liberty of offenders who had already been “tried or punished” for their offences and resulted in harsher penalties than they would have received under the legislation that was in force at the time of their sentencing. Nor does the fact that the legislation extending their parole ineligibility was passed in the context of the CCRA  rather than that of the Criminal Code  shield Parliament’s measure from the scrutiny under s. 11( h )  of the Charter  that would otherwise apply. If that were the case, Parliament could enact punitive laws resulting in double punishment simply by choosing to amend the CCRA  rather than the Criminal Code .

[74]                          I would note that the constitutionality of other aspects of the abolition of APR was not seriously argued. Without a full record, I am not in a position to rule on the effect of changes to the review process (requiring offenders to apply for release, eliminating automatic referral and requiring a hearing rather than a review conducted on paper) and to the release criteria (replacing the lower, presumptive standard of violent recidivism, which left the Board no discretion to deny parole, with the more onerous standard of undue risk to society). These changes are certainly more similar to the change to the sentence management process that was upheld under s. 7 in Cunningham. This being said, every retrospective change must be analyzed in detail before conclusions can be drawn as to its possible punitive effect. The greater the impact on the offender’s settled expectation of liberty, or the greater the likelihood of additional incarceration, the more likely it is that a given retrospective change will violate s. 11(h).

(4)         Does Section 10(1) of the AEPA Violate Section 7 of the Charter ?

[75]                          The respondents argue, in the alternative, that s. 10(1) violates s. 7  of the Charter  by depriving the offender of liberty in contravention of the principles of fundamental justice. Wilson J. made the following comment in this regard in R. v. Gamble, [1988] 2 S.C.R. 595, at p. 647:

The principles of fundamental justice are to be found “in the basic tenets of our legal system”: Re B.C. Motor Vehicle Act, [[1985] 2 S.C.R. 486,] at p. 503. It is fundamental to any legal system which recognizes “the rule of law” (see the Preamble to the Charter ) that an accused must be tried and punished under the law in force at the time the offence is committed. 

Having already found that the impugned provision offends s. 11( h )  of the Charter , I find it unnecessary to decide on the application of s. 7.

[76]                          When both s. 7 and a specific guarantee under the Charter  are pleaded, this Court has generally shown a preference for dealing with the specific guarantee: R. v. Harrer, [1995] 3 S.C.R. 562, at para. 13. In my view, the following comments from R. v. Généreux, [1992] 1 S.C.R. 259, though made in respect of s. 11(d) rather than s. 11(h), are relevant:

The appellant places reliance upon both s. 11( d )  and s. 7  of the Charter . However, the s. 7 submission can be dealt with very briefly. In Re B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486, this Court decided that ss. 8  to 14  of the Charter , the “legal rights”, are specific instances of the basic tenets of fairness upon which our legal system is based, and which are now entrenched as a constitutional minimum standard by s. 7. Consequently, in the context of the appellant’s challenge to the independence of the General Court Martial before which he was tried, s. 7 does not offer greater protection than the highly specific guarantee under s. 11(d). I do not wish to be understood to suggest by this that the rights guaranteed by ss. 8  to 14  of the Charter  are exhaustive of the content of s. 7, or that there will not be circumstances where s. 7 provides a more compendious protection than these sections combined. However, in this case, the appellant has complained of a specific infringement which falls squarely within s. 11(d), and consequently his argument is not strengthened by pleading the more open language of s. 7. [Emphasis added; p. 310.]

 

In R. v. Pearson, [1992] 3 S.C.R. 665, at p. 688, this Court applied the same reasoning in the context of a s. 11(e) claim. Section 11(h) protects against the specific infringement alleged in the case at bar: the retrospective legislative imposition of delayed parole eligibility, which has the effect of punishing the offender anew. I therefore decline to address the arguments with respect to s. 7.

B.            Is Section 10(1) of the AEPA Saved by Section 1  of the Charter ?

[77]                           The trial judge held, and the Court of Appeal agreed, that the objectives of s. 10(1) of the AEPA are pressing and substantial, but they both found that the provision failed at the minimal impairment stage. I agree with this result.

[78]                          As I mentioned above, Parliament based its decision to abolish APR on considerable evidence, presented by the Crown in this case, that the system was not working effectively. It was within Parliament’s prerogative to pass legislation it thought necessary to improve the system. The trial judge found that the objective of ensuring that sentences as administered are consistent with the sentences courts impose, which, by extension, includes maintaining or restoring public confidence in the administration of justice, is pressing and substantial (para. 121). Parliament’s decision to apply these same objectives to all offenders, including those who had already been sentenced, reflects its legitimate concern to ensure uniformity of parole administration and maintain confidence in the justice system. Having agreed that these objectives are pressing and substantial, I also find that the legislative measure — the AEPA — chosen by Parliament, including the retrospective application of its provisions, is rationally connected to the objectives.

[79]                          However, the Crown has not discharged its burden of proving that there was no less intrusive alternative to retrospective application of the AEPA’s provisions.  Uniformity of parole administration may be a worthy objective, but the Crown has failed to provide compelling evidence that that uniformity would be impaired if the APR system continued to apply to offenders who were sentenced under it. I adopt the Court of Appeal’s conclusion in this regard:

Sentence management objectives in general, and the objectives of the AEPA in particular, are recognizably important, but they do not rise to such significance that justifies implementing them in a manner that deprives the respondents of their constitutional rights.  The corrections authorities have for twenty years administered different parole regimes for different offenders, including APR. [para. 65]

[80]                          In my view, having the repeal apply only prospectively was an alternative means available to Parliament that would have enabled it to attain the objectives of reforming parole administration and maintaining confidence in the justice system without violating the s. 11(h) rights of offenders who had already been sentenced. Regarding the Crown’s argument that retrospective application is necessary to maintain confidence in the justice system, I would point out that the enactment of Charter-infringing legislation does great damage to that confidence. The Crown has produced no evidence to show why the alternative of a prospective repeal, which would have been compatible with the respondents’ constitutional rights, would have significantly undermined its objectives.

C.            Did the Court of Appeal Order the Appropriate Remedy?

[81]                          The Crown argues in this Court that the remedy ordered by the summary trial judge and subsequently upheld by the Court of Appeal was overly broad, because she declared s. 10(1) to be invalid in its entirety, whereas only the delay in day parole eligibility — one consequence of the retroactive application of the repeal — had been found to infringe the respondents’ rights.  The Crown submits that the offenders should benefit not from all aspects of the abolished APR system, including automatic referral to the Board, a review conducted on paper and more favourable release criteria, but only from the one aspect — early day parole eligibility — whose elimination has been found to be unconstitutional.

[82]                          Both of the courts below held that the retrospective application of the elimination of early day parole eligibility was unconstitutional. No serious analysis was conducted into the constitutionality of the retrospective application of the elimination of other aspects of the APR system. The “punishment” identified by the summary trial judge and upheld by the Court of Appeal was the delayed date for day parole eligibility, not the changes to the review procedure or to the criteria for release.

[83]                          Although I will not decide this issue, the retrospective modification of the review procedure and the release criteria would seem to me to resemble the change relating to sentence management that this Court accepted in Cunningham. The following comments of McLachlin J., although made in the context of s. 7  of the Charter , are on point:

 A change in the form in which a sentence is served, whether it be favourable or unfavourable to the prisoner, is not, in itself, contrary to any principle of fundamental justice. Indeed, our system of justice has always permitted correctional authorities to make appropriate changes in how a sentence is served, whether the changes relate to place, conditions, training facilities, or treatment. Many changes in the conditions under which sentences are served occur on an administrative basis in response to the prisoner’s immediate needs or behaviour. Other changes are more general. From time to time, for example, new approaches in correctional law are introduced by legislation or regulation. These initiatives change the manner in which some of the prisoners in the system serve their sentences. [Emphasis added; pp. 152-53.]

 

In the instant case, I have held that the legislative imposition of delayed day parole eligibility, by effectively prolonging the minimum period of incarceration of offenders who had already been sentenced, constitutes “punishment” for the purposes of s. 11(h). I have not found other aspects of the AEPA to be unconstitutional. At first blush, I do not see anything impractical or offensive about having the pre-APR parole procedures that are now in force apply while at the same time following the timelines for early day parole eligibility. 

[84]                          However, the courts below did not receive submissions on the limited declaration now being proposed in this Court. This Court must exercise caution in deciding whether to depart from the remedy crafted by the trial judge and upheld by the Court of Appeal. In my view, this cautious approach favours upholding the full declaration of invalidity. I draw this conclusion for two reasons. 

[85]                          First, early day parole eligibility was adopted as a component of APR, and I cannot presume that Parliament would have enacted the legislation establishing it in the absence of the rest of the APR system. The test for severance is whether the constitutionally sound portion of the scheme is so inextricably bound up with the part declared invalid that what remains cannot independently survive (Schachter v. Canada, [1992] 2 S.C.R. 679, at p. 697). I harbour doubts about the respondents’ submission that early day parole was functionally bound up with the rest of the APR scheme. However, to conclude that it was not in the absence of evidence on how early day parole would work in combination with normal parole procedures would be to overstep my judicial role. I cannot presume that Parliament would have enacted the legislation establishing early day parole on its own.

[86]                          Second, even if it were clear that early day parole eligibility could stand, functionally, on its own, severance would not be workable in the context of this legislation. Section 119.1, which established the earlier eligibility period for day parole, was expressly dependent on both s. 125, which defined the class of offenders who were automatically eligible for accelerated parole review, and s. 126, which set out the procedural and substantive terms for release on parole. Section 119.1 read as follows:

119.1 The portion of the sentence of an offender who is eligible for accelerated parole review under sections 125 and 126 that must be served before the offender may be released on day parole is six months, or one sixth of the sentence, whichever is longer.

Section 126.1 specified that ss. 125 and 126 applied to the early day parole option established by s. 119.1.

[87]                          Section 10(1) of the AEPA made the repeal of ss. 125 , 126  and 126.1  of the CCRA  apply retrospectively and thus, in effect, withdrew the right to be considered for early day parole from offenders who had been eligible for such consideration prior to the enactment of the repeal provisions. The specific reference in s. 119.1 to “the sentence of an offender who is eligible for accelerated parole review under sections 125 and 126” means that this section cannot operate without ss. 125, 126 and 126.1 and is thus meaningless without them.

[88]                          To read s. 10(1) of the AEPA down to make the repeal of ss. 125 to 126.1 apply retrospectively but not that of s. 119.1, as the Crown proposes, would result in a provision that would be impossible to implement. I would therefore uphold the declaration of invalidity of all of s. 10(1).

V.           Disposition

[89]                          I would dismiss the appeal and uphold the remedy ordered by the trial judge. Section 10(1) of the AEPA violates s. 11( h )  of the Charter  and is accordingly of no force or effect. Sections 125, 126, 126.1 and, by implication, 119.1 of the CCRA  therefore continue to apply to offenders who were sentenced prior to the coming into force on March 28, 2011, of the AEPA.

                    Appeal dismissed.

                    Solicitor for the appellant:  Attorney General of Canada, Vancouver and Montréal.

                    Solicitors for the respondents:  Eric Purtzki, Vancouver; Garth Barriere, Vancouver.

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

                    Solicitors for the intervener the British Columbia Civil Liberties Association:  University of British Columbia, Vancouver; Heenan Blaikie, Vancouver.