Hyperlawhypertext legal database La Presse inc. v. Quebec 2023 SCC 22

← Cases

La Presse inc. v. Quebec

2023 SCC 22 Court SCC Decision date 2023-10-06 Heard 2023-05-16 Docket 40175, 40223 English

PartiesLa Presse inc. Appellant·His Majesty The King Respondents·Frédérick Silva Respondents

Counsel (4)
Solicitors for the appellant La Presse inc.: Prévost Fortin D’Aoust, Boisbriand, Que.
Solicitor for the respondent His Majesty The King (40175): Director of Criminal and Penal Prosecutions, Québec.
Solicitor for the respondent His Majesty The King (40223): B.C. Prosecution Service — Criminal Appeals and Special Prosecutions, Victoria.
Solicitors for the respondent Aydin Coban: Martland & Saulnier, Vancouver.

Judges Wagner, Richard; Karakatsanis, Andromache; Côté, Suzanne; Martin, Sheilah; Kasirer, Nicholas; Jamal, Mahmud; O’Bonsawin, Michelle

Delivered by Wagner C.J. (Karakatsanis, Côté, Martin, Kasirer, Jamal and O’Bonsawin JJ. concurring)

On appeal from Quebec

Subjects Criminal law

Contents

Reasons for Judgment — Wagner C.J. (Karakatsanis, Côté, Martin, Kasirer, Jamal and O’Bonsawin JJ. concurring)

I. Introduction

In 1972, Parliament enacted an automatic publication ban that prohibits the publication of information about portions of a criminal trial at which the jury is not present. Today, this ban is found in s. 648(1) of the Criminal Code , R.S.C. 1985, c. C‑46 :
648 (1) After permission to separate is given to members of a jury under subsection 647(1), no information regarding any portion of the trial at which the jury is not present shall be published in any document or broadcast or transmitted in any way before the jury retires to consider its verdict.
The question before this Court is whether and, if so, how this automatic publication ban applies before the jury is empanelled, given the jurisdiction conferred by s. 645(5) of the Criminal Code upon trial judges, since 1985, to deal with certain matters before the empanelment of the jury:
(5) In any case to be tried with a jury, the judge before whom an accused is or is to be tried has jurisdiction, before any juror on a panel of jurors is called pursuant to subsection 631(3) or (3.1) and in the absence of any such juror, to deal with any matter that would ordinarily or necessarily be dealt with in the absence of the jury after it has been sworn.
In the cases under appeal, numerous matters were dealt with before the empanelment of the jury. In Mr. Silva’s case, these included a Garofoli application and an application for a stay of proceedings for abuse of process. In Mr. Coban’s case, these included a constitutional challenge to another publication ban under s. 486.4(3) of the Criminal Code . Certain media outlets (the appellants before this Court) applied for orders or declarations that would allow the publication of information from the hearings on those matters. The judges in both cases dismissed the media applications, concluding that s. 648(1) applies before the empanelment of the jury. Information from the hearings could not be published until the juries retired for deliberations or were dismissed.
This Court has addressed s. 648(1) in two cases, although neither of them resolves the interpretive issue in these appeals. In R. v. Brassington, 2018 SCC 37, [2018] 2 S.C.R. 616, footnote 1, Justice Abella noted the diverging approaches to this issue. In R. v. J.J., 2022 SCC 28, para. 283, Justice Brown, dissenting in part, treated s. 648(1) as applying before jury selection to information that would ordinarily be dealt with in the absence of a jury, though he provided little analysis. These appeals call upon this Court to resolve this interpretive issue in light of the provision’s text, context, and purpose — the context including the open court principle and the right to a fair trial.
The open court principle has been recognized by this Court as fundamental throughout the entirety of criminal proceedings, that is, both at the “pre-trial” or pre‑empanelment stage and during the trial (Vancouver Sun (Re), 2004 SCC 43, [2004] 2 S.C.R. 332, at para. 27, citing Attorney General of Nova Scotia v. MacIntyre, [1982] 1 S.C.R. 175, at pp. 183 and 186). I pause at this point to note that the term “pre-trial” in English, and the terms “avant le procès”, “préalable au procès”, and “antérieur au procès” in French, have at times been used in the jurisprudence to refer to the period before a jury has been empanelled (see, e.g., R. v. Malik, Bagri and Reyat, 2002 BCSC 80; R. v. Stobbe, 2011 MBQB 293, 277 Man. R. (2d) 65; R. v. Twitchell, 2010 ABQB 692, 509 A.R. 131; La Presse inc. v. Silva, 2022 QCCS 881; R. v. Bebawi, 2019 QCCS 594; R. v. Farhan, 2000 CanLII 18876 (Que. Sup. Ct.); and R. v. Bissonnette, 2021 QCCS 3856, 74 C.R. (7th) 70). Other times, judges have been careful to refer to this period as “pre-jury-selection” or “before the jury is empanelled” in English, and “avant la sélection du jury”, “préalable à la sélection du jury” or “avant la constitution du jury” in French (see, e.g., R. v. Cheung, 2000 ABQB 905, [2001] 3 W.W.R. 713; Canadian Broadcasting Corp. v. Millard, 2015 ONSC 6583, 338 C.C.C. (3d) 227; R. v. Emms, 2012 SCC 74, [2012] 3 S.C.R. 810; R. v. Ouellette, [1998] R.J.Q. 2842 (Sup. Ct.); and R. v. Talon, 2006 QCCS 3031). The latter expressions are more accurate, but I will occasionally use the term “pre-trial” when I refer to others’ reasoning.
Publication bans like the one imposed by s. 648(1) are limitations on court openness that can protect the right of the accused to, and society’s interest in, a fair trial (see, e.g., Dagenais v. Canadian Broadcasting Corp., [1991] 3 S.C.R. 835, p. 879). However, this Court has recognized that the absence of a publication ban can also advance trial fairness: for example, by preventing perjury, “prevent[ing] state and/or court wrongdoing by placing the criminal justice process under public scrutiny”, and encouraging individuals to come forward with relevant new information after hearing about a case (p. 883).
There is no irreconcilable conflict between the open court principle and trial fairness. They both serve to instill public confidence in the justice system. The public can understand the work of the courts, and thus come to trust the judicial process and its outcomes, only if informed of “what a judge decides” and “why the particular decision is made” (Toronto Star Newspapers Ltd. v. Canada, 2010 SCC 21, [2010] 1 S.C.R. 721, at para. 65 (emphasis in the original)). Needless to say, the media play a crucial role in making this possible (Sherman Estate v. Donovan, 2021 SCC 25, at para. 30, citing Khuja v. Times Newspapers Ltd., [2017] UKSC 49, [2019] A.C. 161, at para. 16; Edmonton Journal v. Alberta (Attorney General), [1989] 2 S.C.R. 1326, at pp. 1339-40). The protection of fair trial interests, such as the right to an independent, impartial, and representative jury, is also essential to public confidence in the administration of justice (R. v. Chouhan, 2021 SCC 26, at para. 12, citing R. v. Sherratt, [1991] 1 S.C.R. 509, at pp. 523-24; see also R. v. Kokopenace, 2015 SCC 28, [2015] 2 S.C.R. 398, at para. 55, citing Sherratt, at pp. 523-25, and R. v. Church of Scientology (1997), 33 O.R. (3d) 65 (C.A.), at pp. 118-20).
Here, Parliament has chosen to impose a temporary publication ban for the purposes of shielding the jury from information it has never been permitted to consider and promoting efficient trials.
I conclude that s. 648(1) applies before the jury is empanelled to matters dealt with pursuant to s. 645(5). This conclusion follows from an understanding of the text of s. 648(1) when considered in its full context and in light of Parliament’s purpose. This interpretation does not expand the coverage of the publication ban: only matters that were captured by the ban prior to the enactment of s. 645(5) continue to be captured by it today. This interpretation has not “evolved” or “changed” in a way that departs from any previous meaning held by s. 648(1). The context of modern trials simply reveals s. 648(1)’s full temporal scope.

II. Judgments Below

A. La Presse inc. v. Silva, 2022 QCCS 881

The accused, Mr. Silva, was charged with four counts of murder and one count of attempted murder. During pre-empanelment proceedings, an application for a stay of proceedings and a Garofoli application were brought in relation to the police techniques used to locate and arrest the accused. David J. dismissed both applications and made orders pursuant to s. 648(1) prohibiting the publication and broadcasting of his decisions. (It is anomalous that these “orders” were made given that, when s. 648(1) applies, it applies automatically, by operation of statute.)
La Presse inc. (an appellant in this Court) brought an application to lift the publication bans, relying on Bebawi, for the view that s. 648(1) applies only after the jury is empanelled. David J. dismissed the application on the basis that, on his interpretation, s. 648(1) applies both before and after the jury is empanelled. In the alternative, and regardless of s. 648(1)’s temporal scope, David J. would have upheld the publication bans under the test set out in Dagenais, in R. v. Mentuck, 2001 SCC 76, [2001] 3 S.C.R. 442, and in Sherman.
David J. listed four reasons for his interpretation of s. 648(1). First, the interpretation of s. 648(1) as applying both before and after the jury is empanelled best aligns with the purpose of this provision, namely to ensure that “the pretrial proceedings do not contaminate the fairness of the later trial” (para. 26 (CanLII), quoting Millard, at para. 25). Second, the current practice in criminal proceedings is to deal with many applications before the jury is empanelled. Third, s. 648(1) must be read together with other Criminal Code provisions establishing publication bans for matters dealt with before the jury is empanelled. Fourth, the temporary nature of s. 648(1) bans strikes a fair balance between the protection of freedom of information and the protection of trial fairness.
Following a guilty verdict on the last count against the accused, the orders made under s. 648(1) were lifted.

B. R. v. Coban, 2022 BCSC 880

The accused, Mr. Coban, was charged with several offences relating to child pornography, extortion, child luring, and harassment (see 2022 BCSC 1810). The underlying facts drew national and international attention (2022 BCSC 14, 420 C.C.C. (3d) 114; A.F., CBC et al., para. 8). Numerous pre-empanelment proceedings occurred over a 15-month period, including a constitutional challenge to s. 486.4(3) of the Criminal Code (see 2022 BCSC 14; A.F., CBC et al., at para. 9). The judge was of the view that the automatic publication ban found in s. 648(1) applied to the information relating to the constitutional challenge.
The Canadian Broadcasting Corporation and other media outlets (appellants in this Court, to which I will refer to collectively as “CBC”) applied for a declaration that the s. 648(1) ban applies only after the jury is empanelled and therefore did not prohibit the publication of information about the constitutional challenge. In her reasons, the judge followed the Supreme Court of British Columbia’s decision in Malik, in which it had been held that s. 648(1) extends to “pre-trial proceedings that take place pursuant to s. 645(5), before a jury has been empanelled”, and she dismissed the application (2022 BCSC 880, at para. 6 (CanLII), citing Malik).

III. Jurisdiction and Mootness

Both appeals reached this Court, with leave, directly from the judgment of a superior court by virtue of s. 40 of the Supreme Court Act , R.S.C. 1985, c. S-26 (see Dagenais, at pp. 861-62 and 872).
Both appeals are also moot because, at this point, neither presents a live controversy. The trials have concluded and s. 648(1) no longer prohibits the publication of any information from them. This Court was aware of the possibility of mootness when it granted leave to appeal.
The considerations in Borowski v. Canada (Attorney General), [1989] 1 S.C.R. 342, suggest that this Court should exercise its discretion to hear and decide the appeals, even though they are moot (see also Doucet-Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3, at paras. 16-22). The parties have provided a clear adversarial context, presenting their arguments capably and with dedication. The operation of s. 648(1) results in automatic publication bans that are often of brief duration relative to the typical timeline of an appeal to this Court, and the issue of its proper interpretation is one that is “capable of repetition, yet evasive of review” (Borowski, at p. 364). Given the judicial division on this issue across the country without the possibility of appellate guidance other than from this Court, the concern for judicial economy justifies resolving the issue, especially considering the importance of the rights and interests in play. Lastly, resolution of this issue calls for straightforward statutory interpretation, a task well within the institutional competence of the Court.

IV. Issues in These Appeals

Trial courts are divided on the interpretation of s. 648(1), as previously noted by this Court in Brassington, at para. 4, fn. 1. Some courts have held that s. 648(1) applies only after the jury is empanelled (Cheung; Twitchell; Bebawi; R. v. Wright, 2020 ONSC 7049, 472 C.R.R. (2d) 296). Others have held that s. 648(1) also applies before the jury is empanelled. Of those holding that s. 648(1) applies to matters dealt with before the jury is empanelled, some have found that it applies to all information about all such matters (R. v. Stanley, 2018 SKQB 27). Others have found that it applies only to certain kinds of hearings (R. v. Sandham (2008), 248 C.C.C. (3d) 543 (Ont. S.C.J.); Stobbe) or have read down the phrase “no information” such that only information that would be prejudicial to the accused is captured by s. 648(1) when it applies before the jury is empanelled (R. v. Regan (1997), 159 D.L.R. (4th) 350 (N.S.S.C.); Malik; R. v. Pickton, 2005 BCSC 836; R. v. Valentine (2009), 251 C.C.C. (3d) 120 (Ont. S.C.J.)).
This judicial divide presents two issues:

(a) Does s. 648(1) apply before the jury is empanelled?

(b) If s. 648(1) applies before the jury is empanelled, what hearings and what information are captured by a publication ban under this section?

Interpreting s. 648(1) under the modern approach to statutory interpretation reveals that the provision applies before the jury is empanelled to prohibit the publication of any information from hearings held pursuant to the jurisdiction provided under s. 645(5).

V. Analysis

A. Principles of Statutory Interpretation

It is well established that, under the modern approach to statutory interpretation, “the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament” (E. A. Driedger, Construction of Statutes (2nd ed. 1983), at p. 87). Confusion as to what this might entail in practice endures, despite the apparent simplicity of Driedger’s influential words. For the sake of clarity, I will restate two principles that seem to be at the heart of this confusion.
First, the plain meaning of the text is not in itself determinative and must be tested against the other indicators of legislative meaning — context, purpose, and relevant legal norms (R. v. Alex, 2017 SCC 37, [2017] 1 S.C.R. 967, at para. 31). The apparent clarity of the words taken separately does not suffice because they “may in fact prove to be ambiguous once placed in their context. The possibility of the context revealing a latent ambiguity such as this is a logical result of the modern approach to interpretation” (Montréal (City) v. 2952-1366 Québec Inc., 2005 SCC 62, [2005] 3 S.C.R. 141, at para. 10).
Second, a provision is only “ambiguous” in the sense contemplated in Bell ExpressVu Limited Partnership v. Rex, 2002 SCC 42, [2002] 2 S.C.R. 559, if its words can reasonably be interpreted in more than one way after due consideration of the context in which they appear and of the purpose of the provision (paras. 29-30). This is to say that there is a “real” ambiguity — one that calls for the use of external interpretive aids like the principle of strict construction of penal laws or the presumption of conformity with the Canadian Charter of Rights and Freedoms — only if differing readings of the same provision cannot be decisively resolved through the contextual and purposive approach set out by Driedger (ibid.).
With these principles in mind, I turn now to the interpretive exercise.

B. Text

For ease of reference, I will reproduce the text of s. 648(1) again:
648 (1) After permission to separate is given to members of a jury under subsection 647(1), no information regarding any portion of the trial at which the jury is not present shall be published in any document or broadcast or transmitted in any way before the jury retires to consider its verdict.
According to the appellants, the opening words of s. 648(1), “[a]fter permission to separate is given to members of a jury”, represent a “condition precedent” that restricts the scope of the prohibition imposed by this provision. More specifically, CBC argues that to interpret s. 648(1) as applying before the jury is empanelled would amount to “striking words of limitation from a statutory provision” (A.F., CBC et al., at para. 77 (emphasis added)). La Presse, for its part, argues that such an interpretation would require adding words (i.e. “before or”) to the provision (A.F., La Presse, para. 50).
This interpretation may seem plausible when the opening words of s. 648(1) are read in isolation. But look just ahead to the words that create the prohibition: “. . . no information regarding any portion of the trial at which the jury is not present shall be published . . . .” These additional words already begin to pull one away from the interpretation urged by the appellants.
In my view, the opening phrase “[a]fter permission to separate is given to members of a jury” simply describes the time at which, when it was enacted in 1972, the prohibition would have had any practical value. In 1972, because there were no pre‑empanelment proceedings, the only time when jurors could have received information about a portion of the trial from which they were absent was when they were permitted to separate.
In any case, to the extent that the wording of the provision is plain, an interpretation based on plain meaning alone is not determinative and “cannot prevail if it is at odds with the purpose and context” (Alex, at para. 33). Here, the context and purpose reveal the alternative, and correct, interpretation.

C. Context

Section 648 was introduced into the Criminal Code alongside s. 647 by the Criminal Law Amendment Act, 1972, S.C. 1972, c. 13. Section 647 expanded the trial judge’s ability to permit the jury to separate. Before the 1972 amendment, the judge could not permit the jury to separate where the accused was charged with an offence punishable by death; in such cases, when the jurors were excluded from the courtroom, they were sequestered under the charge of an officer of the court. After the 1972 amendments, the judge could permit the jury to separate no matter the charged offence. For the first time, jurors were able to go home during portions of capital offence trials. Given that such trials were more likely to attract heightened media scrutiny, s. 648 can be seen as enabling this expanded ability for the jury to separate.
The general understanding is that in 1972, the common law precluded a trial judge from making evidentiary rulings until after a jury was empanelled (R. v. Curtis (1991), 66 C.C.C. (3d) 156 (Ont. C.J. (Gen. Div.)), at pp. 157-58; House of Commons Debates, vol. I, 1st Sess., 33rd Parl., December 20, 1984, at p. 1414 (Parliamentary Secretary to Minister of Justice, at second reading of Bill C-18); Duhamel v. The Queen, [1984] 2 S.C.R. 555, at p. 560). This appears to be a corollary flowing from three premises: (1) the common law principle that a trial did not begin, and a judge was not seized with the trial, until the accused was placed in the charge of the jury (Morin v. The Queen (1890), 18 S.C.R. 407, at p. 413); (2) the principle that “[evidentiary] rulings by one judge do not bind another judge who may later deal with the same matter” (Curtis, at p. 158 (emphasis in original), citing Duhamel; see also D. Macdougall, “Continuity of Judicial Rulings After a Mistrial” (2004), 15 C.R. (6th) 273; R. v. Cliche, 2010 QCCA 408); and (3) the fact that an evidentiary ruling made outside of trial could be shielded from review due to the rule against collateral attack (see, e.g., R. v. Litchfield, [1993] 4 S.C.R. 333; R. v. Garofoli, [1990] 2 S.C.R. 1421).
Judges have traced the increasing complexity of criminal trials to the introduction, in 1974, of provisions relating to the admissibility of wiretap evidence (see Curtis, at pp. 157-58; Protection of Privacy Act, S.C. 1973-74, c. 50). The volume and scope of “pre-trial” applications grew further with the proclamation of the Charter (Malik, at para. 21 (CanLII)). Juries were sent home for days or weeks during the hearing of applications prior to the presentation of evidence at trial. Section 645(5) was enacted in response to the evolution of criminal trials and the increasing burdens on the justice system (see Criminal Law Amendment Act, 1985, R.S.C. 1985, c. 27 (1st Supp.), s. 133 (adding s. 645(5) to the Criminal Code); Malik, at para. 21; Cliche, at paras. 36‑37 (CanLII)).
Section 645(5) allowed the trial judge to hear and adjudicate such applications before the jury was empanelled. Concerns expressed in the parliamentary debates included ensuring respect for the jurors’ time and comfort as well as reducing the costs and resources needed to keep a jury sequestered (see House of Commons Debates, at pp. 1391 and 1414).
Further flexibility was introduced in 2011 with the enactment of the Fair and Efficient Criminal Trials Act, S.C. 2011, c. 16. These provisions, as amended, create the role of a case management judge, who, “before the stage of the presentation of the evidence on the merits . . . exercises the powers that a trial judge has before that stage” (Criminal Code , s. 551.3(1)). When the case management judge adjudicates in relation to matters listed in s. 551.3(1)(g) (evidence, Charter applications, expert witnesses, severance of counts, or separation of trials), the judge’s decision is normally “binding on the parties for the remainder of the trial” (s. 551.3(4)).
The result is that in a modern trial, the bulk of so-called “pre-trial” applications are dealt with by the trial judge or case management judge before the jury is empanelled.
Section 648(1) operates alongside numerous other provisions establishing publication bans, particularly ss. 517(1), 539(1) and 542(2), and the inherent jurisdiction of a judge to impose a discretionary ban under the Dagenais/Mentuck/Sherman framework. Section 517 allows a judge to prohibit the publication of “the evidence taken, the information given or the representations made and the reasons, if any, given or to be given” at a bail hearing. Section 539 allows a judge to prohibit the publication of “the evidence taken” at the preliminary inquiry. The bans in ss. 517 and 539 are mandatory when requested by the accused but discretionary when requested by the Crown. Section 542(2) creates an automatic publication ban with respect to “any admission or confession [that] was tendered in evidence at a preliminary inquiry”. Discretionary bans are those that may be ordered at the discretion of the court; mandatory bans are those that must be imposed at the request of a particular party, and automatic bans are bans that apply by operation of statute. See generally J. Rossiter, Law of Publication Bans, Private Hearings and Sealing Orders (loose-leaf), at §§ 1:7 and 4:48-4:58.
I note that one of the decisions interpreting s. 648(1) as applying only after the jury is empanelled — Bebawi — relied on the principle set out in Orphan Well Association v. Grant Thornton Ltd., 2019 SCC 5, [2019] 1 S.C.R. 150, that “the plain meaning of a provision [cannot] be contorted to make its scheme more coherent” (para. 101). In Orphan Well, a majority of this Court was of the view that the context of a provision in federal bankruptcy legislation confirmed the interpretation flowing from the plain meaning of its text. The majority rejected an alternate interpretation which, even if it might have enhanced the coherence of the scheme, would have impermissibly been at odds with the plain meaning. Here, the context of s. 648(1) does not confirm the interpretation that might arise from an initial impression of the text.
In Bebawi, the judge’s undue reliance on the statement from Orphan Well quoted above led him to give insufficient consideration to the context of s. 648(1), which is a necessary part of the analysis to determine its true meaning. That statement does not necessitate prioritizing the plain meaning of the text to the detriment of the cohesiveness of a statutory scheme. Cohesiveness is a relevant factor, provided that the interpretation to which it leads is one that the words can reasonably bear. Simply put, the interpretive exercise is “incomplete” if it omits a full consideration of one or more of the indicators of legislative meaning, whether it be the text, context, purpose, or relevant legal norms (Alex, at para. 31). In this consideration, [translation] “coherence and logic in a statute . . . provide one argument among others, whose weight varies with the circumstances, and, in particular, with the nature of the statute being interpreted” (P.-A. Côté and M. Devinat, Interprétation des lois (5th ed. 2021), at para. 1120). While the judge in Bebawi acknowledged the role of context and purpose (at paras. 44-45), with respect, his interpretation unduly privileged the plain meaning of the text. Therefore, the reasoning and interpretation in Bebawi should not be followed.
Here, to understand the operation of s. 648(1), one must interpret it in light of the numerous relevant provisions that followed its enactment, and most particularly s. 645(5). In this context, trial judges now have the flexibility to hear, before the empanelment of the jury, various matters that are deemed to be part of the trial. These are clearly dealt with in the absence of the jury and, as such, are automatically covered by s. 648(1).
In situating s. 648(1) in this context, however, I am not suggesting that there is perfect coherence among the various amendments grafted onto the Criminal Code over time. Despite its form as “codal” legislation, the Criminal Code is “not a Code in the civil law sense” and is “less precise than many others” (A. W. Mewett, “The Criminal Law, 1867-1967” (1967), 45 Can. Bar Rev. 726, at pp. 730-31). There has been no systematic revision, and amendments have been described as “haphazard and inconsistent” and as amounting to “isolated tinkering” (A. W. Mewett, “Criminal Law Revision in Canada” (1969), 7 Alta L. Rev. 272, at pp. 277-78). This Court has been careful not to insist on a uniform interpretation of words across the Criminal Code (see, e.g., R. v. Hibbert, [1995] 2 S.C.R. 973, at p. 997).

D. Purpose

By enacting s. 648(1) in 1972, Parliament intended to enhance trial fairness through the protection of two interconnected interests. First, there is the fundamental interest of the accused in being tried by jurors who are not exposed to, and biased by, the rulings rendered on matters dealt with in their absence. I find this to be immediately apparent from the wording of the provision — which bans the publication of information regarding portions of the trial at which the jury is not present — and readily inferable from Hansard. Second, trial fairness under s. 648(1) is also concerned with the interest of both the accused and society in the efficiency of our system of trial by jury. This is revealed by Parliament’s choice to introduce an automatic publication ban that applies simply by operation of statute and thus does not require the intervention of a court.
These two interests are best served when the trial proceeds only on information properly available to the jury. I elaborate on each in turn below.

(1) Right to an Impartial Jury

All parties agree that s. 648(1) was designed to protect the accused’s right to an impartial jury, which in 1972 was part of the common law and which was later constitutionalized under s. 11(f) of the Charter (see N. R. Hasan, “Three Theories of ‘Principles of Fundamental Justice’” (2013), 63 S.C.L.R. (2d) 339, at p. 354). Section 648(1) protects this interest by providing a safeguard against the very risk of jury contamination that came with the introduction of s. 647 of the Criminal Code . Under the latter provision, as mentioned above, judges were allowed for the first time to permit jurors to separate in capital offence trials, which often attracted considerable media attention.
The parties disagree, however, on whether Parliament intended s. 648(1) to act as a safeguard against the contamination of the jury generally or whether it was concerned strictly with the empanelled jury.
The appellants submit that the former view improperly assumes that Parliament would not have viewed the factors that distinguish the pre- and post-jury empanelment stages as warranting different treatment. Rather, the appellants say, Parliament must have intended s. 648(1) to apply only during the “critical” period after the jury is empanelled because of (1) the plethora of mechanisms available before and during selection — challenges for cause, the judge’s instructions to the jury, the swearing-in of the jury, etc. — to guarantee the impartiality of the jurors; (2) the lower potential for prejudice before the jury is selected, given that many matters dealt with at that stage do not involve information prejudicial to the accused; and (3) the sometimes long period between the laying of charges and the empanelment of the jury. According to the appellants, these factors would have led Parliament to adopt a blanket prohibition applicable only at the post-empanelment stage, thereby leaving pre-empanelment publication bans to be discretionary.
I disagree with the appellants. While I recognize that the periods before and after the jury is empanelled are distinguishable by the above-cited features, I see no indication that Parliament turned its mind to such features when it enacted s. 648(1). In reality, Parliament would have had no reason to do so because, as I have said, jurors in 1972 were at risk of being exposed to information about matters dealt with in their absence only when they separated.
I note that s. 648(1)’s purpose was formulated broadly in the parliamentary debates, with concerns being expressed about the publicity of matters dealt with in the absence of the jury. When the publication ban was introduced in 1972, it was noted in the House of Commons Standing Committee on Justice and Legal Affairs that the provision would ensure “that those matters remain a secret for better functioning of jurors” (Minutes of Proceedings and Evidence of the Standing Committee on Justice and Legal Affairs, No. 7, 4th Sess., 28th Parl., May 11, 1972, at p. 7:26). And in the Standing Senate Committee on Legal and Constitutional Affairs, it was explained that s. 648(1) would “prohibit publication of those things that go on when the jury is out of the courtroom” (Proceedings of the Standing Senate Committee on Legal and Constitutional Affairs, No. 8, 4th Sess., June 1, 1972, at p. 8:18). The implication was that, without such a provision, “the members of the jury could go home at night, and would know exactly what went on in the courtroom” (ibid.). There was no other discussion of the publication ban.
On careful consideration of the provision’s wording and the above excerpts from Hansard, I am led to the conclusion that one of Parliament’s objectives was to shield the jury from information about any portion of the trial from which it was absent, so that its verdict is based only on the evidence found admissible in court (see similar expressions of purpose in R. v. Brown (1997), 72 C.R.R. (2d) 312 (Ont. C.J. (Gen. Div.)), at pp. 319-21; Regan; R. v. Bernardo, [1995] O.J. No. 247 (QL), 1995 CarswellOnt 7200 (WL) (C.J. (Gen. Div.)), at para. 43 (QL); Stobbe, at para. 13; Millard, at para. 25; Wright, at para. 25).
This objective is relevant with respect to both the existent jury and the prospective jury, that is, the jury yet to be empanelled. This does not mean that the purpose of s. 648(1) has “shifted” in response to changing practices in criminal proceedings. Justice Dickson (as he then was) rejected the theory of a shifting purpose: “Purpose is a function of the intent of those who drafted and enacted the legislation at the time, and not of any shifting variable” (R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 335). It is simply that, as a result of the introduction of s. 645(5), the matters that should “remain a secret” for the jury are now dealt with both before and after its empanelment.

(2) Interest in an Efficient System of Trial by Jury

Parliament in 1972 intended to protect another interest falling within the right to a fair trial, namely, the interest of both the accused and society in the efficiency of our system of trial by jury. Parliament made information which would otherwise have been covered by discretionary publication bans subject to an automatic publication ban and, in so doing, demonstrated a clear concern for speedy trials and the proper expenditure of judicial resources.
Historically, the media were expected not to prejudice judicial proceedings, such that “the publication of improper information before a case is heard or the dissemination of improper information about a case which is to be heard or is not fully heard” was punishable as contempt of court (see Steiner v. Toronto Star Ltd., [1956] O.R. 14 (H.C.J.), at p. 20, quoting with approval R. v. Evening Standard Co. Ld., [1954] 1 Q.B. 578, at p. 584; see also St. James’s Evening Post Case (1742), 2 Atk. 469, 26 E.R. 683). In other words, publication could be punished, at the remedy stage, even in the absence of an express publication ban. Eventually, it became the practice for courts to issue publication bans as an exercise of their inherent jurisdiction (see, e.g., R. v. Jansen, [1976] 4 W.W.R. 277 (B.C.S.C.); Scott v. Scott, [1913] A.C. 417 (H.L.); R. v. Clement (1821), 4 B. & Ald. 218, 106 E.R. 918).
Under s. 648(1), information about a specific subset of matters which would otherwise have been prohibited from publication by way of discretionary bans was made subject to an automatic publication ban. The latter requires no action to be taken by the parties and the judge; it applies automatically when the stipulated conditions are met. Clearly, Parliament must have had delays and judicial resources in mind when it removed judicial discretion relating to the publication of information about matters dealt with in the absence of the jury.
An absurd interpretation of s. 648(1) that would defeat this objective of efficiency must be avoided. Authors Côté and Sullivan note, respectively, that the label of absurdity can be attached to interpretations that are “incompatible” with the object of the legislative enactment or that “defeat the purpose” of the statute in question (P.‑A. Côté, The Interpretation of Legislation in Canada (2nd ed. 1991), at pp. 378-80, and R. Sullivan, Driedger on the Construction of Statutes (3rd ed. 1994), at p. 88, both cited in Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27, at para. 27).
While no evidence has been provided to that effect, one can reasonably expect that an interpretation confining the application of s. 648(1) to the post‑empanelment stage would lead to a multiplication of applications for discretionary Dagenais/Mentuck/Sherman bans. This, in turn, would most likely result in further delays in the criminal justice system and the diversion of scarce resources of the accused and the court. Such a result would be antithetical to the objective of efficiency pursued by Parliament in enacting s. 648(1) and completely at odds with the teachings of this Court in R. v. Jordan, 2016 SCC 27, [2016] 1 S.C.R. 631.
Conversely, an interpretation of s. 648(1) as applying before the jury is empanelled can be expected to further the efficiency interest. Plainly, by shielding information arising at many types of hearings from publication, s. 648(1) gives courts the confidence to hold such hearings prior to the empanelment of the jury. This flexibility and ability to hold such hearings earlier in time can be expected to reduce delays. It may also allow the parties to gain certainty about contested matters, such as the admissibility of evidence, in advance of when rulings on those matters would historically have been made, leading to earlier resolution by way of pleas or withdrawal of charges.

(3) Conclusion

All of the above leads me to conclude that s. 648(1) was designed to safeguard the right to a fair trial by averting jury bias and by ensuring the efficiency of our system of trial by jury. This is consistent with this Court’s existing conception of trial fairness as being concerned not only with averting jury bias by banning “pre-trial” publicity but also with protecting the accused’s other fundamental interests. In Toronto Star, at para. 23, Justice Deschamps characterized Parliament’s primary objectives in enacting the publication ban in s. 517 of the Criminal Code as being “(1) to safeguard the right to a fair trial; and (2) to ensure expeditious bail hearings”. The former, she wrote, embraces the latter (para. 24). A similar view was adopted in Jordan, where fair trial interests were said to be affected by delays in the criminal justice system. This is so because “the longer a trial is delayed, the more likely it is that some accused will be prejudiced in mounting a defence, owing to faded memories, unavailability of witnesses, or lost or degraded evidence” (para. 20). More recently, in R. v. Haevischer, 2023 SCC 11, at para. 46, this Court built on the Jordan vocabulary and presented trial fairness and trial efficiency as “interdependent” values.

E. Meaning of Section 648(1)

All indicators of legislative meaning — text, context, and purpose — admit of only one interpretation of s. 648(1): that it applies not only after the jury is empanelled but also before the jury is empanelled with respect to matters dealt with pursuant to s. 645(5). Said another way, if a judge would traditionally have had jurisdiction to hear a matter before the jury was empanelled without having to rely on s. 645(5), then that matter is not within the scope of s. 648(1). As I have found no ambiguity after conducting the full interpretive exercise, there is no need, contrary to what some of the parties argued, to resort to external aids such as the principle of strict construction of penal laws or the presumption of conformity with the Charter (Bell ExpressVu, at para. 55).
In 1972, information heard at trials in the absence of the jury was prohibited from publication. The law prohibiting the publication of such information has not changed now that those hearings also happen before the jury is empanelled. Without clear language, “a statute should not be interpreted as substantially changing the law” (R. v. D.L.W., 2016 SCC 22, [2016] 1 S.C.R. 402, at para. 21). Section 648 has not been substantially amended, and s. 645(5) does not contain any language signifying an intent to remove hearings held pursuant to s. 645(5) from the scope of the publication ban.
In closing, I will say a word on the use of subsequent legislative history, that is, the record of the process, materials, and debate that followed s. 648(1)’s enactment, in the interpretation of this provision.
The appellants in both appeals make much of the withdrawal of a clause that would have amended s. 648(1) to make it expressly applicable “in respect of any matter dealt with by a judge before any juror is sworn”. In 1994, the Minister of Justice explained to the House of Commons that the government’s initial intention was to “fill a gap” with respect to evidence-related pre-empanelment motions whose publication “might contaminate members of a prospective jury” (House of Commons Debates, vol. 133, No. 143, 1st Sess., 35th Parl., December 13, 1994, at p. 9010 (emphasis added)). He added that the language of clause 62 “might be overbroad” and that it was accordingly to be removed from Bill C-42 (ibid.).
In the submission of La Presse and CBC, the fact that Parliament considered but ultimately refrained from expressly broadening s. 648(1)’s scope indicates that this provision has always — from 1972 until today — been meant to apply only after the jury is empanelled. Put differently, they ask this Court to infer that because Parliament in 1994 might have found it necessary to amend s. 648(1), this tells us something about what Parliaments in 1972 or 1985 intended with respect to this provision and its interaction with s. 645(5).
I would decline to place any weight on the 1994 amending clause and its withdrawal. My view is that proposed but abandoned amendments are of no assistance in identifying the meaning of the legislation that would otherwise have been amended. In this regard, I adopt the reasoning of Justices Cory and Iacobucci in M. v. H., [1999] 2 S.C.R. 3, at para. 105:
With respect, I cannot agree that a failed amendment can provide evidence as to the objective of the legislation that was to have been amended. Section 17 of the Interpretation Act, R.S.O. 1990, c. I.11, provides: “The repeal or amendment of an Act shall be deemed not to be or to involve any declaration as to the previous state of the law.” If the amendment of an Act may not be used to interpret the meaning of the Act prior to the amendment, then I do not see how a failed amendment may be used in this manner. [Emphasis in original.]

This reasoning applies equally to federal legislation given s. 45(3) of the Interpretation Act , R.S.C. 1985, c. I-21 (stating that the “repeal or amendment of an enactment in whole or in part shall not be deemed to be or to involve any declaration as to the previous state of the law”) and in light of logical limits of relevance.

VI. Determining Whether s. 645(5) Captures a Pre-Empanelment Proceeding

Section 648(2) makes it a summary conviction offence to violate the s. 648(1) publication ban — a ban that applies automatically, by operation of statute, with no need for a court order.
I recognize that interpreting s. 648(1) as applying before the jury is empanelled, but only to some matters, could give rise to uncertainty over what matters are covered by the publication ban (see Millard, at para. 65; I.F., British Columbia Civil Liberties Association, at para. 37). And I am mindful that courts should avoid creating uncertainty, especially in criminal law, because “[i]t is a fundamental requirement of the rule of law that a person should be able to predict whether a particular act constitutes a crime at the time he commits the act” (R. v. Breault, 2023 SCC 9, at para. 27, quoting R. v. Mabior, 2012 SCC 47, [2012] 2 S.C.R. 584, at para. 14). While these concerns did not drive the above interpretive exercise, I wish to provide some guidance to mitigate any uncertainty.
Section 648(2) creates a “true crime” and thus is subject to the presumption that a person should not be held liable unless the person acted intentionally or recklessly, with knowledge of the facts constituting the offence (R. v. A.D.H., 2013 SCC 28, [2013] 2 S.C.R. 269, at para. 23; R. v. Sault Ste. Marie, [1978] 2 S.C.R. 1299, at pp. 1303 and 1309-10). For the practical effectiveness and clarity of the publication ban itself, and to avoid potential barriers to prosecution for violating the ban, it would therefore be prudent for judges holding a hearing pursuant to s. 645(5) to announce that they are exercising their jurisdiction under that section and to note that s. 648(1) automatically prohibits the publication of any information regarding that portion of the trial.
I anticipate that the above approach will provide sufficient certainty. It would not be a worthwhile effort to attempt to provide a comprehensive list of matters that would be captured or excluded by s. 648(1). I note, just as a descriptive observation, that most kinds of hearings taking place before a jury is empanelled will be covered by this prohibition. For example, there is no dispute that evidentiary voir dires would be covered (Duhamel, at p. 560; R. v. Lalo, 2002 NSSC 21, 207 N.S.R. (2d) 203, at para. 19; R. v. Ross, [1995] O.J. No. 3180 (QL), 1995 CarswellOnt 3173 (WL) (C.J. (Gen. Div)), at para. 3, per Salhany J.; R. E. Salhany, Canadian Criminal Procedure (5th ed. 1989), at pp. 189-90): these are matters for which a judge clearly relies on s. 645(5) jurisdiction in order to hold the hearing before the jury is empanelled. There are, however, other kinds of hearings that have never been required to take place “at trial”; these would not be covered by the prohibition found in s. 648(1). This determination can be made through the analysis outlined in Litchfield.
This Court’s analysis in Litchfield provides a useful framework for assessing whether a matter is being dealt with by virtue of s. 645(5) or whether it could always have been dealt with, even in the absence of s. 645(5), before the jury was empanelled. In Litchfield, the question was whether a motion for division or severance of counts had to be made “at trial” or if it could be dealt with before the trial by a judge other than the trial judge. This Court looked to the following features in holding that division and severance motions must be dealt with “at trial” and can therefore be dealt with only by virtue of s. 645(5) if the jury has not yet been empanelled:

(a) Does the motion “concer[n] the indictment”? Must the indictment have been “preferred” or, in the language of R. v. Chabot, [1980] 2 S.C.R. 985, does the indictment need to be the “operative document”? See Litchfield, at pp. 350-52.

(b) Must the person hearing the matter be the trial judge seized with the trial? This includes a consideration of practical or policy reasons to have the matter dealt with by the trial judge (see Litchfield, at pp. 352-53). Today, this question would be asked in the context of the new jurisdiction of case management judges under s. 551.3: But for the jurisdiction of case management judges, would the matter have to be dealt with by the trial judge?

In Litchfield, these features together led to the conclusion that a motion for division or severance of counts had to be dealt with by the trial judge. For one thing, the majority concluded that “no one except the trial judge ever has jurisdiction to divide or sever counts since an indictment is only preferred at the opening of an accused’s trial” (p. 352). Further, the practical or policy rationale was that if it were open to a judge other than the trial judge to rule on these motions, the resulting order would be immunized from review due to the rule against collateral attack: “. . . ‘a court order, made by a court having jurisdiction to make it,’ may not be attacked ‘in proceedings other than those whose specific object is the reversal, variation, or nullification of the order or judgment’” (p. 348, quoting Wilson v. The Queen, [1983] 2 S.C.R. 594, at p. 599).
Justice Iacobucci, writing for the majority, continued:
. . . it makes sense that the trial judge consider applications to divide and sever counts since an order for division or severance of counts will dictate the course of the trial itself. . . . Not only are trial judges better situated to assess the impact of the requested severance on the conduct of the trial, but limiting severance orders to trial judges avoids the duplication of efforts to become familiar enough with the case to determine whether or not a severance order is in the interests of justice. It seems desirable, therefore, that in the future only trial judges can make orders for division or severance of counts in order to avoid injustices such as occurred in this case. [Emphasis added; p. 353.]
Concerns about appealability also played a role in Garofoli. There, the majority held that it is the trial judge, rather than the issuing court, that must review a wiretap authorization when the accused asserts an infringement of s. 8 of the Charter and asks that the resulting evidence be suppressed under s. 24 (pp. 1448-49). The majority first noted that “[a]n objection to the reception of evidence is very much a necessary incident of the trial process” (p. 1449), and also found that placing this review in the hands of the trial judge “would remove any doubts as to the right to appeal a decision rejecting or admitting the evidence” (p. 1450). Sopinka J. was skeptical that such a decision would be reviewable on appeal under the former approach, where evidence would be admitted or rejected by the trial judge based on the order of a “Wilson judge” (p. 1450).
This reasoning from Litchfield and Garofoli highlights the sorts of considerations that have drawn courts to conclude that certain matters must be dealt with by the trial judge (again, but for the new jurisdiction of case management judges).
To address submissions from the Crown respondents on the pre‑empanelment scope of s. 648(1), I wish to make two further observations in relation to case management judges and pre-trial conferences.
The Quebec Crown submitted that the list of matters in s. 551.3(1)(g) could help clarify the pre‑empanelment scope of s. 648(1). Sections 551.1 to 551.7 were introduced as Part XVIII.1 of the Criminal Code by the Fair and Efficient Criminal Trials Act, s. 4. These provisions, as amended, allow a case management judge to “exercis[e] the powers that a trial judge has before” “the stage of the presentation of the evidence on the merits” (s. 551.3(1)). Section 551.3(1)(g) does not expand the powers of a case management judge beyond those of a trial judge. Section 551.3(1)(g) allows the case management judge to “adjudicat[e] any issues that can be decided before that stage”. Thus, even under s. 551.3(1)(g), some of what a case management judge does is traceable to the jurisdiction that a trial judge only has under s. 645(5) to hear a matter that would ordinarily or necessarily be dealt with in the absence of the jury after it has been sworn (see M. Vauclair and T. Desjardins, in collaboration with P. Lachance, Traité général de preuve et de procédure pénales 2022 (29th ed. 2022), at p. 587, fn. 117). When a case management judge exercises jurisdiction traceable to s. 645(5), s. 648(1) automatically applies.
The B.C. Crown submitted that s. 648(1) covers even the pre-trial conference required for jury trials by s. 625.1(2). Section 625.1 was introduced alongside s. 645(5) in 1985 and came into force in 1988. Section 625.1(2) mandates a conference, not a hearing of matters. These conferences are held under rules enacted by courts of criminal jurisdiction and provincial and territorial courts under ss. 482 and 482.1 of the Criminal Code . They are generally for considering matters that would promote a fair and expeditious trial, including the scheduling of hearings, and that can be resolved before the jury is empanelled (see, e.g., Criminal Rules of the Supreme Court of British Columbia, SI/97-140, r. 5; Rules of Practice of the Superior Court of the Province of Quebec, Criminal Division, 2002, SI/2002-46, rr. 39 to 44). Discussion at the pre-trial conference will include information such as the “nature and particulars” of proposed motions (B.C. Rules, r. 5(13)), and statements relating to the “respective positions of the parties” (Que. Rules, r. 44(b)).
I reiterate that s. 648(1) applies before the jury is empanelled only when a judge is exercising jurisdiction traceable to s. 645(5) to deal with a matter that would ordinarily or necessarily be dealt with in the absence of the jury after it has been sworn. Generally, a pre-trial conference is not that. The Alberta Court of Justice, however, has enacted a rule allowing the pre-trial conference judge to “make any ruling a case management judge acting under section 551.3 of the Code may make, except a ruling under sections 551.3(1)(e) or 551.3(1)(g) of the Code” (Criminal Rules of the Alberta Court of Justice, r. 4.2(7)(a)). Without deciding whether rules enacted under ss. 482 and 482.1 can expand the jurisdiction of a pre-trial conference judge,[1] I emphasize that s. 648(1) would apply to information from a pre-trial conference only when the pre-trial conference judge exercises jurisdiction ultimately rooted in s. 645(5). The Alberta Court of Justice seems to have excluded several such matters from adjudication at a pre-trial conference.
Publication of information that comes out at pre-trial conferences may very well be prejudicial to the fair trial interests of the accused. The fact that there will be potentially prejudicial proceedings not covered by the automatic publication ban found in s. 648(1) was recognized as far back as 1979. In R. v. Deol (1979), 20 A.R. 595 (Q.B.), at para. 31, the judge noted that s. 648(1) (then s. 576.1) left a “lamentable gap” in relation to “possibly prejudicial proceedings between the end of the preliminary hearing and the moment when the jury is first permitted to separate”. But s. 648(1), as interpreted above, does not reach so far as to apply automatically to all aspects of a pre‑trial conference. This would be an interpretation that the text cannot reasonably bear. It is of course open to courts to fill any gap in relation to pre-trial conferences through their rule-making authority under ss. 482 and 482.1 for example, the Superior Court of Quebec has made it a rule that “[p]roceedings at the pre-hearing conference are subject to a publication ban” (Que. Rules, r. 40) and judges retain inherent jurisdiction to impose discretionary publication bans in accordance with the Dagenais/Mentuck/Sherman principles.

VII. Dispositions

Since I have interpreted s. 648(1) as applying before the jury is empanelled to matters dealt with pursuant to s. 645(5), I would dismiss both appeals.
In Mr. Silva’s case, David J. understood s. 648(1) as applying to matters dealt with before the jury is empanelled. He did not specify, however, whether it would cover information about all matters or only those dealt with pursuant to s. 645(5). Regardless, the order dismissing the application to lift the publication bans should be upheld. One of the matters dealt with by David J. concerned the admissibility of evidence (a Garofoli application). The other was a motion for a stay of proceedings for abuse of process. This clearly concerned the indictment and had to be dealt with by the trial judge (or case management judge exercising the powers of a trial judge) so that it would be reviewable on appeal from the conviction. Therefore, it is only by virtue of s. 645(5) that these matters could be dealt with prior to the empanelment of the jury, and it follows that they were covered by s. 648(1). I would dismiss the appeal in this case.
The reasoning in Mr. Coban’s case was that s. 648(1) applies to “all pre‑trial applications” (para. 2). This is not consistent with the proper interpretation of s. 648(1). However, the order was simply as follows: “. . . the Application to clarify or declare that the publication ban herein pursuant to section 648 of the Criminal Code of Canada applies only to proceedings after the jury is empanelled, is dismissed” (A.R., CBC et al., at pp. 7-8). The media had applied for a declaration that s. 648(1) applies only after the jury has been empanelled. The judge dismissed that application. That was the extent of the order. While the judge did not adopt the interpretation I have presented, the dismissal is consistent with the proper interpretation of s. 648(1). I would therefore dismiss the appeal in this case as well.
I would award no costs as none were sought.

Footnotes

  1. [1] This issue was considered in R. v. Commanda, 2007 QCCA 947, [2008] 3 C.N.L.R. 311, at para. 47, and R. v. S. (S.S.) (1999), 136 C.C.C. (3d) 477 (Ont. S.C.J.), at para. 47, per Watt J. Both rejected the possibility that rules enacted under s. 482 can confer jurisdiction on a judge when this jurisdiction has not been given by statute.

Cases cited

Statutes and regulations cited