Toronto Star Newspapers Ltd. et al. v. Her Majesty the Queen in Right of Canada et al.; N.Y. (being a Young Person within the meaning of the [Youth Criminal Justice Act](https://www.canlii.org/en/ca/laws/stat/sc-2002-c-1/latest/sc-2002-c-1.html)) et al., Interested Parties [Indexed as: Toronto Star Newspapers Ltd. v. Canada]

94 O.R. (3d) 82

103 O.R. (3d) 399

Court of Appeal for Ontario,

Laskin, Rosenberg, Feldman, Simmons and Juriansz

JJ.A.

January 26, 2009

Charter of Rights and Freedoms -- Freedom of the press -- Publication ban -- Mandatory publication ban at bail hearing at request of accused infringing s. 2(b) of Charter -- Section 1 test not met as s. 517 of Code overly broad as applicable to all charges regardless of whether could be tried by jury -- Appropriate remedy being to read down s. 517 to apply only to charges that could be tried by jury -- Canadian Charter of Rights and Freedoms, ss. 1, 2(b) -- Criminal Code, R.S.C. 1985, c. C-46, s. 517.

Criminal law -- Bail -- Publication ban -- Mandatory publication ban at request of one accused under s. 517 applying to bail hearings of all jointly charged accused even if another accused opposes the ban -- Criminal Code, R.S.C. 1985, c. C-46, s. 517.

Criminal law -- Appeal -- Mootness -- Mandatory publication ban at request of accused under s. 517 overtaken by discretionary ban by time of appeal -- Discretionary ban not contested -- Constitutionality of s. 517 still contested between Crown and media -- Outcome of case will have practical effect on parties and appeal raising important issues that will recur -- Time limits on publication bans making it likely that future bans would be over before litigation of constitutionality could reach [page83 ]appellate court -- Court exercising discretion to determine moot appeal -- Criminal Code, R.S.C. 1985, c. C-46, s. 517.

A number of accused were charged with terrorism-related offences. One of the accused requested a ban on publication of proceedings at the bail hearings pursuant to s. 517 of the Criminal Code, which requires a judge to impose a mandatory publication ban where the accused requests it. Despite the fact that some of the accused did not want a publication ban, the justice of the peace imposed a ban. On applications by media organizations to quash the bans, the application judge held that when one of several jointly charged accused seeks the mandatory order under s. 517, it applies to the bail hearings of all the accused. In a separate decision, he upheld the constitutionality of the mandatory publication ban. By the time that the appeal was heard, the earlier mandatory ban had been supplanted by a discretionary ban which was not contested by the media. The media organizations appealed.

Held, the appeal should be allowed in part.

Per Feldman J.A. (Laskin and Simmons JJ.A. concurring): It was appropriate to hear the appeal even though it was moot. This case has the unusual feature that some of the accused attack the constitutionality of the provision while others seek to uphold it, giving the court the benefit of broader submissions than many s. 2(b) cases in which only the media and the government are active participants. As s. 517 bans are time limited, a publication ban would likely expire in most cases before a party contesting its constitutionality could obtain an appellate ruling. This appeal raises an important legal issue that will recur. It is appropriate to exercise the court's discretion to hear a moot appeal.

The mandatory publication ban in s. 517 violates s. 2(b) of the Canadian Charter of Rights and Freedoms. It is not saved by s. 1 because it is overly broad in its scope applying to bail hearings for all charges, including those that cannot be tried by a jury. The purpose of the publication ban itself is to protect the accused's right to a fair trial by an impartial jury, which is a pressing and substantial goal because of the potential for jury contamination. A mandatory ban protects the accused's right to expeditious bail proceedings which may be impaired if a bail hearing is delayed to give notice to the media or to permit the accused to retain counsel and prepare for a contested hearing on a discretionary ban. These objectives are pressing and substantial.

A publication ban on bail proceedings is rationally connected to the preservation of an accused's right to a fair trial by jury. However, to the extent that it also applies to cases in which a jury trial isn't possible, it is overbroad.

The s. 517 publication ban, and the mandatory nature of the ban, meets the minimal impairment test to the extent that it applies to matters that may be tried by a jury.

The salutary effects of the mandatory publication ban in cases where a jury trial is possible outweigh the deleterious effects. It protects the accused from prejudicial pre-trial publication of evidence that may be not be admissible at trial and its mandatory nature is one of its greatest benefits by preventing delays in bail hearings and avoiding additional costs. The ban isn't permanent and the media and the public can attend bail proceedings despite a publication ban. The limitation on freedom of expression is justified under s. 1 of the Charter only in cases where a jury trial is possible. To the extent that s. 517 applies in respect of charges that cannot be tried by a jury, the legislation is not justified under s. 1. The appropriate remedy is to revise s. 517 by adding the italicized words: "If the [page84 ]prosecutor or the accused intends to show cause under s. 515, he or she shall so state to the justice and the justice may, and shall on application by the accused where and for so long as the charge(s) may be tried by a jury . . .".

Per Rosenberg J.A. (Juriansz J.A. concurring) (dissenting): The mandatory publication ban at the request of the accused under s. 517 of the Criminal Code infringes s. 2(b) of the Charter. The government has not met its s. 1 onus justifying this infringement because the mandatory nature of the ban fails the proportionality test. The deleterious effects of a mandatory ban include a significant restriction on the freedom of expression and preventing an informed public discussion about important legal issues as even the reasons for the release of the accused cannot be made. In addition, it applies in cases in which a jury trial either could not, or is highly unlikely, to occur. The benefits attributed to a mandatory ban are largely speculative and can be achieved by other means. Most bail hearings are dealt with by justices of the peace. In most cases, they could exercise their discretion not to give notice to the media, minimizing risk to the accused's right to an expeditious bail hearing if making a publication ban were to be discretionary. The decision in Global Communications has been overtaken by later judgments and should be overturned. The remedy for this Charter violation is to read down s. 517 by removing the words "shall on application of the accused". However, given the possibility that Parliament may wish to revisit the provision and concerns that justices of the peace may not have jurisdiction to make the kind of order that would meet constitutional standards, the remedy is suspended for one year.

APPEAL from the orders of Durno J., 2006 25418 (S.C.J.) and (2007), 2007 6249 (ON SC), 84 O.R. (3d) 766 (S.C.J.) dismissing applications to quash publication bans.

Cases referred to Global Communications Ltd. and Canada (Attorney General) (Re) (1984), 1984 2153 (ON CA), 44 O.R. (2d) 609, [1984] O.J. No. 3066, 5 D.L.R. (4th) 634, 2 O.A.C. 21, 10 C.C.C. (3d) 97, 38 C.R. (3d) 209, 7 C.R.R. 22, 11 W.C.B. 276 (C.A.), affg (1983), 1983 1851 (ON SC), 42 O.R. (2d) 13, [1983] O.J. No. 3043, 148 D.L.R. (3d) 331, 5 C.C.C. (3d) 346, 34 C.R. (3d) 52, 5 C.R.R. 38, 9 W.C.B. 367 (H.C.J.), ovruld R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, [1986] S.C.J. No. 7, 26 D.L.R. (4th) 200, 65 N.R. 87, 14 O.A.C. 335, 24 C.C.C. (3d) 321, 50 C.R. (3d) 1, 19 C.R.R. 308, 16 W.C.B. 73, apld Dagenais v. Canadian Broadcasting Corp., 1994 39 (SCC), [1994] 3 S.C.R. 835, [1994] S.C.J. No. 104, [1994] 3 S.C.R. 835, 120 D.L.R. (4th) 12, 175 N.R. 1, J.E. 95-30, 76 O.A.C. 81, 94 C.C.C. (3d) 289, 34 C.R. (4th) 269, 25 C.R.R. (2d) 1, 51 A.C.W.S. (3d) 1045, 25 W.C.B. (2d) 304, consd Other cases referred to Borowski v. Canada (Attorney General), 1989 123 (SCC), [1989] 1 S.C.R. 342, [1989] S.C.J. No. 14, 57 D.L.R. (4th) 231, 92 N.R. 110, [1989] 3 W.W.R. 97, J.E. 89-499, 75 Sask. R. 82, 47 C.C.C. (3d) 1, 33 C.P.C. (2d) 105, 38 C.R.R. 232, 7 W.C.B. (2d) 61; Canada (Attorney General) v. JTI-Macdonald Corp., [2007] 2 S.C.R. 610, [2007] S.C.J. No. 30, 2007 SCC 30, 281 D.L.R. (4th) 589, 364 N.R. 89, J.E. 2007-1276, 158 C.R.R. (2d) 127, 158 A.C.W.S. (3d) 919; Canadian Broadcasting Corp. v. New Brunswick (Attorney General), 1996 184 (SCC), [1996] 3 S.C.R. 480, [1996] S.C.J. No. 38, 139 D.L.R. (4th) 385, 203 N.R. 169, 182 N.B.R. (2d) 81, 110 C.C.C. (3d) 193, 2 C.R. (5th) 1, 39 C.R.R. (2d) 189, 66 A.C.W.S. (3d) 444, 32 W.C.B. (2d) 273; Canadian Newspapers Co. v. Canada (Attorney General), 1988 52 (SCC), [1988] 2 S.C.R. 122, [1988] S.C.J. No. 67, 52 D.L.R. (4th) 690, 87 N.R. 163, J.E. 88-1134, 32 O.A.C. 259, 43 C.C.C. (3d) 24, 65 C.R. (3d) 50, 38 C.R.R. 72, 5 W.C.B. (2d) 6; Church of Scientology of Toronto and the Queen (No. 6) (Re), 1986 4649 (ON SC), [1986] O.J. No. 3068, 27 C.C.C. (3d) 193, 16 W.C.B. 449 (H.C.J.); Cochrane v. Ontario (Attorney General) (2008), 92 O.R. (3d) 321, [2008] O.J. No. 4165, 2008 ONCA 718, 242 O.A.C. 192, 170 A.C.W.S. (3d) 508, 61 C.R. (6th) 374; [page85 ]David Polowin Real Estate Ltd. v. Dominion of Canada General Insurance Co. (2005), 2005 21093 (ON CA), 76 O.R. (3d) 161, [2005] O.J. No. 2436, C.C.L.I. (4th) 191, 15 C.P.C. (6th) 1, [2005] I.L.R. I-4422, 19 M.V.R. (5th) 205, 140 A.C.W.S. (3d) 166 (C.A.) [Leave to appeal to S.C.C. refused [2005] S.C.C.A. No. 388]; Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326, [1989] S.C.J. No. 124, 4 D.L.R. (4th) 577, 102 N.R. 321, [1990] 1 W.W.R. 577, J.E. 90-47, 71 Alta. L.R. (2d) 273, 103 A.R. 321, 41 C.P.C. (2d) 109, 45 C.R.R. 1, 18 A.C.W.S. (3d) 894; Gay Alliance Toward Equality v. Vancouver Sun, 1979 225 (SCC), [1979] 2 S.C.R. 435, [1979] S.C.J. No. 55, 97 D.L.R. (3d) 577, 27 N.R. 117, [1979] 4 W.W.R. 118; Harper v. Canada (Attorney General), [2004] 1 S.C.R. 827, [2004] S.C.J. No. 28, 2004 SCC 33, 239 D.L.R. (4th) 193, 320 N.R. 49, [2004] 8 W.W.R. 1, J.E. 2004-1104, 27 Alta. L.R. (4th) 1, 348 A.R. 201, 119 C.R.R. (2d) 84, 130 A.C.W.S. (3d) 746; Hollinger Inc. v. Ravelston Corp. (2008), 89 O.R. (3d) 721, [2008] O.J. No. 1126, 2008 ONCA 207, 164 A.C.W.S. (3d) 771, 235 O.A.C. 136, 291 D.L.R. (4th) 15, 53 C.P.C. (6th) 222, 169 C.R.R. (2d) 136 [Leave to appeal to S.C.C. refused, [2008] S.C.C.A. No. 260]; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927, [1989] S.C.J. No. 36, 58 D.L.R. (4th) 577, 94 N.R. 167, J.E. 89-772, 24 Q.A.C. 2, 25 C.P.R. (3d) 417, 39 C.R.R. 193, 15 A.C.W.S. (3d) 121; Named Person v. Vancouver Sun, [2007] 3 S.C.R. 253, [2007] S.C.J. No. 43, 2007 SCC 43, 285 D.L.R. (4th) 193, 368 N.R. 112, [2008] 1 W.W.R. 223, J.E. 2007-1909, 247 B.C.A.C. 1, 73 B.C.L.R. (4th) 34, 224 C.C.C. (3d) 1, 51 C.R. (6th) 262, 75 W.C.B. (2d) 103, 162 C.R.R. (2d) 104, EYB 2007-124673; Osborne v. Canada (Treasury Board), 1991 60 (SCC), [1991] 2 S.C.R. 69, [1991] S.C.J. No. 45, 82 D.L.R. (4th) 321, 125 N.R. 241, J.E. 91-1000, 37 C.C.E.L. 135, 91 CLLC Â14,026 at 12231, 4 C.R.R. (2d) 30, 27 A.C.W.S. (3d) 601; Phillips v. Nova Scotia (Commission of Inquiry into the Westray Mine Tragedy), 1995 86 (SCC), [1995] 2 S.C.R. 97, [1995] S.C.J. No. 36, 124 D.L.R. (4th) 129, 180 N.R. 1, J.E. 95-945, 141 N.S.R. (2d) 1, 31 Admin. L.R. (2d) 261, 98 C.C.C. (3d) 20, 39 C.R. (4th) 141, 28 C.R.R. (2d) 1, 54 A.C.W.S. (3d) 1115, 27 W.C.B. (2d) 9; R. v. Adams, 1995 56 (SCC), [1995] 4 S.C.R. 707, [1995] S.C.J. No. 105, 131 D.L.R. (4th) 1, 190 N.R. 161, J.E. 96-115, 178 A.R. 161, 103 C.C.C. (3d) 262, 44 C.R. (4th) 195, 29 W.C.B. (2d) 185; R. v. B., [2006] E.W.C.A. Crim. 2692, [2007] E.M.L.R. 145 (C.A. Crim. Div.); R. v. Banville, 1983 3027 (NB QB), [1983] N.B.J. No. 110, 145 D.L.R. (3d) 595, 45 N.B.R. (2d) 134, 3 C.C.C. (3d) 312, 34 C.R. (3d) 20, 5 C.R.R. 142, 9 W.C.B. 102 (Q.B.); R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, [1985] S.C.J. No. 17, 18 D.L.R. (4th) 321, 58 N.R. 81, [1985] 3 W.W.R. 481, 37 Alta. L.R. (2d) 97, 60 A.R. 161, 18 C.C.C. (3d) 385, 85 CLLC Â14,023 at 12108, 13 C.R.R. 64, 14 W.C.B. 157; R. v. Brooks, 2001 28401 (ON SC), [2001] O.J. No. 1563, [2001] O.T.C. 306, 153 C.C.C. (3d) 533, 49 W.C.B. (2d) 533 (S.C.J.); R. v. Bryan, [2007] 1 S.C.R. 527, [2007] S.C.J. No. 12, 2007 SCC 12, 276 D.L.R. (4th) 513, 359 N.R. 1, [2007] 5 W.W.R. 1, J.E. 2007-530, 237 B.C.A.C. 33, 72 B.C.L.R. (4th) 199, 217 C.C.C. (3d) 97, 45 C.R. (6th) 102, 153 C.R.R. (2d) 316, 72 W.C.B. (2d) 362; R. v. Burke, [2002] 2 S.C.R. 857, [2002] S.C.J. No. 56, 2002 SCC 55, 213 D.L.R. (4th) 234, 290 N.R. 71, J.E. 2002-1232, 160 O.A.C. 271, 164 C.C.C. (3d) 385, 2 C.R. (6th) 1, 53 W.C.B. (2d) 300; R. v. Butler, 1992 124 (SCC), [1992] 1 S.C.R. 452, [1992] S.C.J. No. 15, 89 D.L.R. (4th) 449, 134 N.R. 81, [1992] 2 W.W.R. 577, J.E. 92-353, 78 Man. R. (2d) 1, 70 C.C.C. (3d) 129, 11 C.R. (4th) 137, 8 C.R.R. (2d) 1, 15 W.C.B. (2d) 159; R. v. Corbett, 1988 80 (SCC), [1988] 1 S.C.R. 670, [1988] S.C.J. No. 40, 85 N.R. 81, [1988] 4 W.W.R. 481, J.E. 88-737, 28 B.C.L.R. (2d) 145, 41 C.C.C. (3d) 385, 64 C.R. (3d) 1, 34 C.R.R. 54, 4 W.C.B. (2d) 148; R. v. Daly, [2005] B.C.J. No. 1697, 2005 BCCA 389, 256 D.L.R. (4th) 372, 215 B.C.A.C. 107, 198 C.C.C. (3d) 185, 66 W.C.B. (2d) 171, affg [2003] B.C.J. No. 1742, 2003 BCSC 1143, 178 C.C.C. (3d) 31, 58 W.C.B. (2d) 518, [2003] B.C.T.C. 1143; R. v. Darrach, [2000] 2 S.C.R. 443, [2000] S.C.J. No. 46, 2000 SCC 46, 191 D.L.R. (4th) 539, 259 N.R. 336, J.E. 2000-1931, 137 O.A.C. 91, 148 C.C.C. (3d) 97, 36 C.R. (5th) 223, 78 C.R.R. (2d) 53, 47 W.C.B. (2d) 360; R. v. Domm (1996), 1996 1331 (ON CA), 31 O.R. (3d) 540, [1996] O.J. No. 4300, 95 O.A.C. 262, 111 C.C.C. (3d) 449, 4 C.R. (5th) 61, 40 C.R.R. (2d) 289, 33 W.C.B. (2d) 108 (C.A.) [Leave to appeal to S.C.C. refused [1997] S.C.C.A. No. 78, 43 C.R.R. (2d) 188]; [page86 ]R. v. Edwards Books and Art Ltd., 1986 12 (SCC), [1986] 2 S.C.R. 713, [1986] S.C.J. No. 70, 35 D.L.R. (4th) 1, 71 N.R. 161, J.E. 87-82, 19 O.A.C. 239, 30 C.C.C. (3d) 385, 87 CLLC Â14,001 at 12001, 55 C.R. (3d) 193, 28 C.R.R. 1, 1 W.C.B. (2d) 19; R. v. Hall, [2002] 3 S.C.R. 309, [2002] S.C.J. No. 65, 2002 SCC 64, 217 D.L.R. (4th) 536, 293 N.R. 239, J.E. 2002-1881, 165 O.A.C. 319, 167 C.C.C. (3d) 449, 4 C.R. (6th) 197, 97 C.R.R. (2d) 189, 54 W.C.B. (2d) 599; R. v. Handy (2002), 61 O.R. (3d) 414, [2002] 2 S.C.R. 908, [2002] S.C.J. No. 57, 2002 SCC 56, 213 D.L.R. (4th) 385, 290 N.R. 1, J.E. 2002-1226, 160 O.A.C. 201, 164 C.C.C. (3d) 481; R. v. Jevons, [2008] O.J. No. 4397, 2008 ONCJ 559; R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697, [1990] S.C.J. No. 131, 117 N.R. 1, [1991] 2 W.W.R. 1, J.E. 91-42, 77 Alta. L.R. (2d) 193, 114 A.R. 81, 61 C.C.C. (3d) 1, 1 C.R. (4th) 129, 3 C.R.R. (2d) 193, 11 W.C.B. (2d) 352; R. v. Kenny, 1991 2738 (NL SC), [1991] N.J. No. 253, 92 Nfld. & P.E.I.R. 318, 68 C.C.C. (3d) 36, 14 W.C.B. (2d) 9 (T.D.); R. v. Lake, [1997] O.J. No. 5446 (Gen. Div.); R. v. Mentuck, [2001] 3 S.C.R. 442, [2001] S.C.J. No. 73, 2001 SCC 76, 205 D.L.R. (4th) 512, 277 N.R. 160, [2002] 2 W.W.R. 409, J.E. 2001-2142, 163 Man. R. (2d) 1, 158 C.C.C. (3d) 449, 47 C.R. (5th) 63, 51 W.C.B. (2d) 349; R. v. Parks (1993), 1993 3383 (ON CA), 15 O.R. (3d) 324, [1993] O.J. No. 2157, 65 O.A.C. 122, 84 C.C.C. (3d) 353, 24 C.R. (4th) 81, 21 W.C.B. (2d) 121 (C.A.) [Leave to appeal to S.C.C. refused (1993), 17 O.R. (3d) xvii, [1993] 1 S.C.R. x, [1993] S.C.C.A. No. 481, 72 O.A.C. 159n, 87 C.C.C. (3d) vi, 1994 19087 (SCC), 28 C.R. (4th) 403n]; R. v. Pearson, 1992 52 (SCC), [1992] 3 S.C.R. 665, [1992] S.C.J. No. 99, 144 N.R. 243, J.E. 92-1760, 52 Q.A.C. 1, 77 C.C.C. (3d) 124, 17 C.R. (4th) 1, 12 C.R.R. (2d) 1, 17 W.C.B. (2d) 576; R. v. Powley, [2003] 2 S.C.R. 207, [2003] S.C.J. No. 43, 2003 SCC 43, 230 D.L.R. (4th) 1, 308 N.R. 201, J.E. 2003-1751, 177 C.C.C. (3d) 193, 5 C.E.L.R. (3d) 1, [2003] 4 C.N.L.R. 321, 110 C.R.R. (2d) 92, 125 A.C.W.S. (3d) 1, 58 W.C.B. (2d) 348, affg (2001), 2001 24181 (ON CA), 53 O.R. (3d) 35, [2001] O.J. No. 607, 196 D.L.R. (4th) 221, 141 O.A.C. 121, 152 C.C.C. (3d) 97, [2001] 2 C.N.L.R. 291, 40 C.R. (5th) 221, 80 C.R.R. (2d) 1, 49 W.C.B. (2d) 193; R. v. Seaboyer, 1991 76 (SCC), [1991] 2 S.C.R. 577, [1991] S.C.J. No. 62, 83 D.L.R. (4th) 193, 128 N.R. 81, J.E. 91-1312, 48 O.A.C. 81, 66 C.C.C. (3d) 321, 7 C.R. (4th) 117, 6 C.R.R. (2d) 35, 13 W.C.B. (2d) 624; R. v. Sharpe, [2001] 1 S.C.R. 45, [2001] S.C.J. No. 3, 2001 SCC 2, 194 D.L.R. (4th) 1, 264 N.R. 201, [2001] 6 W.W.R. 1, J.E. 2001-294, 146 B.C.A.C. 161, 88 B.C.L.R. (3d) 1, 150 C.C.C. (3d) 321, 39 C.R. (5th) 72, 86 C.R.R. (2d) 1, 48 W.C.B. (2d) 287; R. v. Smith, 2006 30593 (Ont. S.C.J.); R. v. Vetrovec, 1982 20 (SCC), [1982] 1 S.C.R. 811, [1982] S.C.J. No. 40, 136 D.L.R. (3d) 89, 41 N.R. 606, [1983] 1 W.W.R. 193, J.E. 82-563, 67 C.C.C. (2d) 1, 27 C.R. (3d) 304, 7 W.C.B. 477; R. v. White, [2005] A.J. No. 1727, 2005 ABCA 435, 56 Alta. L.R. (4th) 255, 376 A.R. 63, 68 W.C.B. (2d) 70; R. v. White, [2006] A.J. No. 179, 2006 ABCA 65, 380 A.R. 188, 69 W.C.B. (2d) 103; R. v. White, [2008] A.J. No. 956, 2008 ABCA 294, 93 Alta. L.R. (4th) 239, [2008] 10 W.W.R. 588, 437 A.R. 130, 236 C.C.C. (3d) 204, 298 D.L.R. (4th) 659, revg 2007 ABQB 359, [2007] A.J. No. 608, [2007] 10 W.W.R. 250, 77 Alta. L.R. (4th) 98, 221 C.C.C. (3d) 393, 48 C.R. (6th) 300, 158 C.R.R. (2d) 270, 420 A.R. 1, 74 W.C.B. (2d) 189 (Q.B.); R. v. Y. (K.S.) (October 11, 2002) (Ont. C.A.); RJR-MacDonald Inc. v. Canada (Attorney General), 1995 64 (SCC), [1995] 3 S.C.R. 199, [1995] S.C.J. No. 68, 127 D.L.R. (4th) 1, 187 N.R. 1, J.E. 95-1766, 100 C.C.C. (3d) 449, 62 C.P.R. (3d) 417, 31 C.R.R. (2d) 189, 57 A.C.W.S. (3d) 578, 28 W.C.B. (2d) 216; Ruby v. Canada (Solicitor General), [2002] 4 S.C.R. 3, [2002] S.C.J. No. 73, 2002 SCC 75, 219 D.L.R. (4th) 385, 295 N.R. 353, J.E. 2002-2096, 49 Admin. L.R. (3d) 1, 22 C.P.R. (4th) 289, 7 C.R. (6th) 88, 99 C.R.R. (2d) 324, 118 A.C.W.S. (3d) 2; Schachter v. Canada, 1992 74 (SCC), [1992] 2 S.C.R. 679, [1992] S.C.J. No. 68, [1992] 2 S.C.R. 679, 93 D.L.R. (4th) 1, 139 N.R. 1, J.E. 92-1054, 92 CLLC Â14,036 at 12221, 10 C.R.R. (2d) 1, 34 A.C.W.S. (3d) 1090; Sierra Club of Canada v. Canada (Minister of Finance), [2002] 2 S.C.R. 522, [2002] S.C.J. No. 42, 2002 SCC 41, 211 D.L.R. (4th) 193, 287 N.R. 203, J.E. 2002-803, 40 Admin. L.R. (3d) 1, 44 C.E.L.R. (N.S.) 161, 20 C.P.C. (5th) 1, 18 C.P.R. (4th) 1; Thomson Newspapers Co. v. Canada (Attorney General) (1998), 1998 829 (SCC), 38 O.R. (3d) 735, [1998] 1 S.C.R. 877, [1998] S.C.J. No. 44, 159 D.L.R. (4th) 385, 226 N.R. 1, J.E. 98-1224, 109 O.A.C. 201, 51 C.R.R. (2d) 189, 79 A.C.W.S. (3d) 921; [page87 ]Toronto Star Newspapers Ltd. v. Ontario, [2005] 2 S.C.R. 188, [2005] S.C.J. No. 41, 2005 SCC 41, 253 D.L.R. (4th) 577, 335 N.R. 201, J.E. 2005-1234, 200 O.A.C. 348, 197 C.C.C. (3d) 1, 29 C.R. (6th) 251, 132 C.R.R. (2d) 178; Vancouver Sun (Re), [2004] 2 S.C.R. 332, [2004] S.C.J. No. 41, 2004 SCC 43, 240 D.L.R. (4th) 147, 322 N.R. 161, [2005] 2 W.W.R. 671, 199 B.C.A.C. 1, 33 B.C.L.R. (4th) 261, 184 C.C.C. (3d) 515, 21 C.R. (6th) 142, 120 C.R.R. (2d) 203, 61 W.C.B. (2d) 216 Statutes referred to An Act to amend the Criminal Code, S.C. 2005, c. 32 Bail Reform Act, S.C. 1970-71-72, c. 37 [in R.S.C. 1970, 2nd Supp., c. 2] Canada Evidence Act, R.S.C. 1970, c. E-10, s. 12 Canada Evidence Act, R.S.C. 1985, c. C-5, s. 12 [as am.] Canadian Charter of Rights and Freedoms, ss. 1, 2, (b), 7, 11(d), (e), (f) Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11, s. 52 Criminal Code, R.S.C. 1970, c. C-37, s. 12 Criminal Code, R.S.C. 1985, c. C-46, Part XVI, ss. 276 [as am.], 515 [as am.], 515(10), 517 [as am.], (1) [as am.], 518 [as am.], 539 [as am.], 542 [as am.], (2) [as am.] Criminal Law Amendment Act 1975, S.C. 1974-75-76, c. 93, s. 48 Privacy Act, R.S.C. 1985, c. P-21, s. 51(2)(a) [as am.] Youth Criminal Justice Act, S.C. 2002, c. 1, s. 110 Authorities referred to Dufraimont, Lisa"Evidence Law and the Jury: A Reassessment" (2008), 53 McGill L.J. 199 Friedland, Martin L., Detention Before Trial: A Study of Criminal Cases Tried in the Toronto Magistrates' Court (Toronto: University of Toronto Press, 1965) Greene, Edith, and Mary Dodge"The Influence of Prior Record Evidence on Juror Decision Making" (1995), 19 Law and Human Behaviour 67 Hogg, Peter, Constitutional Law of Canada, 5th ed., looseleaf (Toronto: Carswell, 2007) Kramer, Geoffrey P., et al."Pretrial Publicity, Judicial Remedies, and Jury Bias" (1990), 14 Law and Human Behaviour 409 Law Reform Commission of Canada, Working Paper 56, Public Media Access to the Criminal Process (Ottawa, 1987) Linden, Allen M."Limitations on Media Coverage of Legal Proceedings: A Critique and Some Proposals for Reform" in Anisman, Philip and Allen M. Linden, eds., The Media, The Courts and the Charter (Toronto: Carswell, 1986) McRuer, James Chalmer, Ontario Royal Commission into Civil Rights, Report Number One, vol. 2 (Ottawa: Queen's Printer, 1968) Ontario Court of Justice 2005 Annual Report, online: www.ontariocourts.on.ca/ocj/en/annualreport/2005.pdf Otto, Amy, et al."The Biasing Impact of Pretrial Publicity on Juror Judgments" (1994) 18 Law and Human Behaviour 453 Ouimet, Roger, Report of the Canadian Committee on Corrections -- Towards Unity: Criminal Justice and Corrections (Ottawa: Queen's Printer, 1969) Pickel, Kerri L."Inducing Jurors to Disregard Inadmissable Evidence: A Legal Explanation Does Not Help" (1995), 19 Law and Human Behaviour 407 Trotter, Justice Gary, The Law of Bail in Canada, 2nd ed. (Toronto: Carswell, 1999) [page88 ]

Paul B. Schabas, Ryder L. Gilliland and Iris Fischer, for appellant Toronto Star Newspapers Ltd. John North, Steve A. Coroza and Marco Mendicino, for respondent Her Majesty the Queen in Right of Canada. M. David Lepofsky and Daniel Guttman, for intervenor Her Majesty the Queen in Right of Ontario. Christopher D. Hicks, for respondents Amin Mohamed Durrani, Yoga Krishnan and Suhaib Mohammed, and interested party N.S. John Norris, for respondent Asad Ansari. Raymond Motee, for respondent Ibrahim Aboud. Rocco Galati, for Ahmad Mustafa Ghany. Royland (Jim) Moriah, for respondent Jahmaal James. David Kolinsky, for respondent Zakaria Amara. Peter Martin, for respondent Shareef Abdelhaleen. Michael A. Moon, for respondent Steven Vikash Chand. Ingrid Grant, for respondent Saad Khalid. Dennis Edney, for respondent Fahim Ahmad.

[1] ROSENBERG J.A. (Dissenting) JURIANSZ J.A. (concurring): -- This appeal from two decisions of Durno J. concerns the interpretation and constitutionality of s. 517 of the Criminal Code, R.S.C. 1985, c. C-46. That section requires the judge or justice of the peace presiding at a bail hearing to impose a ban on publication of the proceedings where the prosecutor intends to show cause why the accused should not be released on an undertaking without conditions and the accused requests the publication ban. In 1984, just two years after the Canadian Charter of Rights and Freedoms came into force, this court upheld the constitutionality of the mandatory publication ban in Global Communications Ltd. and Canada (Attorney General) (Re) (1984), 1984 2153 (ON CA), 44 O.R. (2d) 609, [1984] O.J. No. 3066 (C.A.). The application judge followed Global and upheld the constitutionality of the provision. He also resolved an issue of interpretation of s. 517 holding that where one of several jointly charged accused requests the publication ban, the ban applies to the show cause hearings of all of the accused.

[2] Members of the media appeal theses orders. The respondent Her Majesty the Queen in Right of Canada supported by the intervenor the Attorney General of Ontario and several of the other accused seek to uphold the orders of the application judge. Canada and Ontario also submit that the appeal should [page89 ]be dismissed because it is moot, as the initial publication bans have been overtaken by a discretionary publication ban that is not challenged.

[3] For the following reasons, I would allow the appeal in part and hold that the mandatory publication ban violates s. 2(b) of the Charter and cannot be saved by s. 1, and is therefore of no force and effect. I would suspend the declaration of invalidity for 12 months. The Factual Context

[4] In early June 2006, the police arrested 12 adults and five young persons for terrorism-related offences. The police held press conferences to announce the arrests, which attracted worldwide attention. Between June 3 and June 12, when the first publication ban was imposed, details of the plots were disclosed by the police, by CSIS and by defence counsel. These details were widely reported and included allegations that the accused were plotting to blow up various public buildings such as the Parliament buildings, that the plotters had tons of bomb-making materials and weapons, that they had attended terrorist training camps, that they had links with terrorists organizations in the United States and Europe and that they were inspired by al-Qaeda. Almost 5,000 articles reported on the arrests and the allegations.

[5] On June 12, 2006, Justice of the Peace Currie imposed a ban on publication of proceedings at the bail hearings at the request of one of the accused pursuant to s. 517. Some of the accused did not want a publication ban. The justice ruled that since the accused were jointly charged, the ban applied to the proceedings in relation to all of the accused, even those who did not seek it.

[6] The appellant media organizations applied to the Superior Court of Justice for an order quashing the publication bans, as they related to the accused who did not seek them. This application involved a matter of statutory interpretation. In reasons released July 27, 2006, the application judge held that when one of several jointly charged accused seeks the mandatory order under s. 517, it applies to the bail hearings of all the accused.

[7] The media organizations then brought an application seeking a declaration that the mandatory publication ban violates s. 2(b) of the Charter and should be declared of no force and effect. In reasons released March 1, 2007, the application judge held that he was bound by Global Communications and upheld the validity of s. 517. [page90 ]

[8] The media organizations appeal both decisions of the application judge. The Legislative History

[9] At the time the publication ban was imposed, s. 517 of the Criminal Code provided as follows: [See Note 1 below]

517(1) If the prosecutor or the accused intends to show cause under section 515, he or she shall so state to the justice and the justice may, and shall on application by the accused, before or at any time during the course of the proceedings under that section, make an order directing that the evidence taken, the information given or the representations made and the reasons, if any, given or to be given by the justice shall not be published in any document, or broadcast or transmitted in any way before such time as (a) if a preliminary inquiry is held, the accused in respect of whom the proceedings are held is discharged; or (b) if the accused in respect of whom the proceedings are held is tried or ordered to stand trial, the trial is ended.

(2) Every one who fails without lawful excuse, the proof of which lies on him, to comply with an order made under subsection (1) is guilty of an offence punishable on summary conviction.

[10] Section 517 has its origins in the recommendations of the 1969 Report of the Canadian Committee on Corrections -- Towards Unity: Criminal Justice and Corrections (Ottawa: Queen's Printer, 1969) (the "Ouimet Report"), which recommended that the Criminal Code be amended to provide that the judge or justice make an order prohibiting publication of bail hearing proceedings on application by the accused. However, when the extensive amendments to the Code were enacted as part of the Bail Reform Act, S.C. 1970-71-72, c. 37 [in R.S.C. 1970, 2nd Supp., c. 2], the publication ban was discretionary. Then s. 457.2(1) read as follows:

457.2(1) Where the prosecutor intends to show cause under subsection 457(1) [now s. 515], he shall so state to the justice and the justice may, before or at any time during the course of the proceedings under that section, make an order directing that the evidence taken, the information given or the representations made and the reasons, if any, given or to be given by the justice, shall not be published in any newspaper or broadcast before such time as (a) if a preliminary inquiry is held, the accused in respect of whom the proceedings are held is discharged, or [page91 ] (b) if the accused in respect of whom the proceedings are held is tried or committed for trial, the trial is ended.

[11] In 1976, the Code was amended by the Criminal Law Amendment Act 1975, S.C. 1974-75-76, c. 93, s. 48 to provide that the publication ban is mandatory at the request of the accused and discretionary where the order is sought by the prosecution. There does not appear to have been any debate in Parliament to explain the reasons for the amendment. Global Communications

[12] The decision of this court in Global Communications arose out of a bail hearing in an extradition matter. The extradition judge, at the request of the fugitive and the Attorney General of Canada, had made an order banning publication of the bail proceedings. Global Communications then brought an application to quash the order on the basis that the judge had no jurisdiction to make the order and, in the alternative, that the order infringed s. 2 of the Charter. Linden J., in reasons reported at (1983), 1983 1851 (ON SC), 42 O.R. (2d) 13, [1983] O.J. No. 3043 (H.C.J.), held that the Criminal Code bail provisions, including (now) s. 517, apply to extradition proceedings. Thus, the extradition judge had jurisdiction to make the order. He also held that, while the section violated the guarantee of freedom of the press in s. 2(b), it was a reasonable limit within the meaning of s. 1 of the Charter. [See Note 2 below] On appeal to this court, Thorson J.A., speaking for the court, agreed with the reasons of Linden J.

[13] Linden J.'s decision was founded principally on the theory that the accused's right to a fair trial is paramount to the right of freedom of expression, and on the possibility that publication of bail proceedings could threaten trial fairness. Thus, he wrote as follows, at pp. 20 and 23 O.R.:

I have been satisfied by the respondent that this limit on freedom of the press contained in the Code is a reasonable one. The order banning publication in this matter is temporary. Its scope is limited to the evidence given and representations made at the bail hearing. The order did not deny anyone access to the court-room. Nor was anyone stopped from publishing the result of the hearing. The infringement of the applicant's freedom has, thus, been held to a minimum. [page92 ]

Furthermore, it must be recognized that this order, though interfering with freedom of the press, protects another interest which is also guaranteed by the Charter. That other interest contained in s. 11(d) is the fugitive's right to be "presumed innocent until proven guilty according to law in a fair and public hearing by an independent and impartial tribunal". To allow publication of the banned material may taint the minds of potential jurors with irrelevant, inadmissible or unchallenged evidence, damaging the fugitive's ability to be fairly tried. . . . . .

Our courts are conscious of the vital role the press plays in the administration of justice by exposing all of our processes to necessary public scrutiny. This is generally welcomed, but there are times when the freedom of the press and the concomitant right of the public to know must yield to the even more important right to a fair trial before an impartial tribunal. (Emphasis added)

[14] Thorson J.A. agreed with these reasons. He also took the position that the right to a fair trial is fragile, and that unrestricted publication of bail proceedings would irreparably harm that right. He wrote as follows, at p. 625 O.R.:

Notwithstanding that such an order imposes a measure of prior restraint, the restriction which s. 457.2(1) authorizes to be placed on the media's right to publish or broadcast the evidence given at an accused's bail hearing is, in my opinion, a reasonable one in the interests of ensuring that the subsequent trial of the accused will be a fair one.

The right to a fair trial is a fragile right. It is quite capable of being shattered by the kind of publicity that can attend a bail hearing and, once shattered, it may, like Humpty Dumpty, be quite impossible to put back together again. Often the proceedings at a bail hearing do not attract any particular media notice, but when they do, as they have in this case, the risk of prejudice to the accused in the matter of his or her subsequent trial can be severe in the absence of a mechanism such as that provided in s. 457.2(1) for minimizing that prejudice by means of a time-limited restraint on what can be published or broadcast about the hearing. In my opinion it is no answer at all that the need for such a mechanism might be avoided if only our legal system were otherwise and adopted a different approach altogether to the safeguarding of a fair trial for the accused. As has been said in other words elsewhere by this Court, ultimately our courts must come back to our own free and democratic society in applying the test in this regard which s. 1 of the Charter requires us to apply. (Emphasis added) The Preliminary Issues

Mootness

[15] Canada and Ontario submit that a mootness issue arises in this appeal. Following bail proceedings that resulted in some but not all of the accused being released on bail, the accused appeared for their preliminary inquiries. The presiding judge [page93 ]made orders prohibiting publication of the evidence at the preliminary inquiries pursuant to s. 539 of the Criminal Code and s. 110 of the Youth Criminal Justice Act, S.C. 2002, c. 1. The preliminary inquiry proceedings were terminated when the Crown filed a direct indictment in the Superior Court of Justice. On September 24, 2007, the accused appeared in the Superior Court of Justice. At that time, Dawson J. made a 30-day interim discretionary publication ban order covering all proceedings in the Ontario Court of Justice, including the bail proceedings. The media were given notice that they had 30 days to challenge this order. No member of the media challenged the order and on October 22, 2007, the Superior Court of Justice extended the discretionary order until the end of trial.

[16] The order by Dawson J. prohibiting publication of the proceedings in the Ontario Court of Justice was made in accordance with Dagenais v. Canadian Broadcasting Corp., 1994 39 (SCC), [1994] 3 S.C.R. 835, [1984] S.C.J. No. 104. Canada and Ontario submit that the discretionary order has superseded the mandatory s. 517 orders and therefore the question of the validity of the s. 517 orders is moot. In my view, even if the appeal is moot, this is the type of case where the court ought to exercise its discretion to hear the case.

[17] In Borowski v. Canada (Attorney General), 1989 123 (SCC), [1989] 1 S.C.R. 342, [1989] S.C.J. No. 14, at pp. 358-63 S.C.R., the court listed three factors a court should consider in deciding whether to exercise its discretion to hear a moot appeal: (a) whether an adversarial relationship still exists between the parties; (b) whether special circumstances exist in the case so as to justify the expenditure of scarce judicial resources; and (c) whether there is a need for the court to be sensitive to its role as the adjudicative branch of government.

[18] An adversarial relationship still exists between the appellants and Canada. As was said in Borowski, at p. 359 S.C.R., even if the litigant no longer has a direct interest in the outcome"there may be collateral consequences of the outcome that will provide the necessary adversarial context". Although the initial mandatory orders have been overtaken by the discretionary order, the constitutionality of s. 517 remains a live issue between the parties. Members of the media are keenly interested in having the issue settled and Canada needs to know whether its legislation is valid. The issue will have to be determined at some point [page94 ]and this court has had the benefit of full factums and complete oral argument. Moreover, in the arguments before Dawson J. concerning the discretionary orders, reference was made to the fact that the mandatory bans were already in place and would be frustrated if a discretionary order were not put in place. As he said:

There have also been section 517 orders made in a number of bail hearings and bail reviews. Again, at least some of those orders, if not all, were mandatory pursuant to the provisions of the Criminal Code. They would also be undermined and frustrated.

[19] This case is an excellent vehicle for testing the validity of the provision, not only because of the submissions of the parties, but particularly because of the submissions of the accused. Several of the accused argue against the mandatory publication ban. As the appellants point out, this provides an important dimension that may be missing in a subsequent case.

[20] The second factor, concern for judicial economy, is met where "the special circumstances of the case make it worthwhile to apply scarce judicial resources to resolve it" (Borowski, at p. 360 S.C.R.). One special circumstance is where the court's decision will have some practical effect on the rights of the parties even if it will not determine the controversy that gave rise to the action. This is such a case. As already mentioned, the parties have an interest in the validity of s. 517, since the issue is bound to arise again. As well"expenditure of judicial resources is considered warranted in cases which although moot are of a recurring nature but brief duration" (Borowski, at p. 360 S.C.R.). The question of publication bans at bail hearings will continue to arise. However, the time frame during which bans will be of any practical effect will usually be too brief for the question to reach this court.

[21] This case is not unlike R. v. Adams, 1995 56 (SCC), [1995] 4 S.C.R. 707, [1995] S.C.J. No. 105, 103 C.C.C. (3d) 262, where the court considered the validity of an order lifting a ban on publication of the complainant's name in a sexual assault case. The court held that the case was not moot but indicated that it would have exercised its discretion in any event to hear the case since the issue was important and might affect future cases. In Borowski, at p. 361, the court noted that there exists "a rather ill-defined basis for justifying the deployment of judicial resources in cases which raise an issue of public importance of which a resolution is in the public interest. The economics of judicial involvement are weighed against the social cost of continued uncertainty in the law." This rationale applies in this case. There will continue to be [page95 ]uncertainty about the continuing viability of Global Communications in light of the substantial development of the law by the Supreme Court of Canada since 1984.

[22] Finally, this is not a case where the limits on the court's proper law-making function require that the court not hear the case. The issue raised concerns the validity of court orders made pursuant to a section of the Criminal Code. The case is not a departure from the court's traditional role. Moreover, at the core of the dispute is the continuing viability of one of this court's own decisions. This is pre- eminently a matter for this court to rule on. Accordingly, I would not give effect to the mootness argument.

Fresh evidence

[23] Ontario seeks to introduce fresh evidence of legislative facts to support the constitutionality of the legislation. The media oppose admission of the evidence on the basis that the evidence should have been adduced at the application and is of limited probative value. I would admit the evidence. As this court pointed out at para. 61 of R. v. Powley, (2001), 2001 24181 (ON CA), 53 O.R. (3d) 35, [2001] O.J. No. 607, 152 C.C.C. (3d) 97 (C.A.), affd 2003 SCC 43, [2003] 2 S.C.R. 207, [2003] S.C.J. No. 43, 177 C.C.C. (3d) 193, in constitutional cases appellate courts have allowed considerable latitude for the admission of new materials related to legislative facts. Most of the evidence Ontario seeks to adduce is not controversial and helps to provide some context for the submissions. Some of the information is admittedly not necessary, since the application judge, who had been an experienced defence counsel, has presented a helpful picture of the operation of bail hearings in his reasons, and the practice and procedure at bail hearings is fully reviewed in Justice Gary Trotter's text, The Law of Bail in Canada, 2nd ed. (Toronto: Carswell, 1999). I am therefore satisfied that the appellants are in no way prejudiced by the fresh evidence.

[24] The fresh evidence consists of affidavits and transcripts of cross-examination on those affidavits. The affidavits were provided by two experienced Crown counsel and the former head of Legal Aid Ontario's province-wide criminal duty counsel programme. The evidence shows that duty counsel conduct most of the bail hearings in the province, and that they often prepare for and conduct them under difficult circumstances. The accused have usually been newly arrested. They have limited understanding of the court system and the charges facing them, and limited ability to provide instructions to counsel. A significant number of accused may have mental health or addiction issues. Duty counsel operate under serious time constraints, [page96 ]and facilities for interviewing and providing information to the accused are far from ideal. In short, counsel have to convey and obtain a great deal of information in a very short period, at a time when the accused is particularly vulnerable. Defence counsel is rarely in a position to contest the truth of the allegations made by Crown counsel during the bail hearing.

[25] The evidence shows that there is no uniform approach adopted by counsel in seeking a s. 517 order. Crown counsel rarely seek the discretionary order. Some duty counsel routinely apply for the mandatory order, while others seek the order only where there is some media attention or matters of an especially private nature may be disclosed.

[26] Most criminal charges are disposed of in the Ontario Court of Justice without a jury trial. In fact, approximately 80 per cent of charges are resolved by a guilty plea. That said, on occasion, cases that might appear to be within the exclusive jurisdiction of the Ontario Court of Justice lead to more serious charges upon further investigation; these charges could be the subject of a jury trial.

The interpretation issue

[27] In his 2006 ruling, the application judge held that where one of several jointly charged accused applies for a non- publication order under s. 517, the order applies to the bail proceedings with respect to all of the accused, even though there may be separate bail hearings. This issue arose not simply because the media sought to publish the proceedings, but because several of the accused opposed the non-publication order.

[28] I do not find it necessary to engage in an extensive analysis of this issue since I agree with the reasons of the application judge. At paras. 84 to 148, he conducted a thorough analysis of the issue. The section is reasonably capable of the application judge's interpretation and any other interpretation would undermine the objectives of the legislation.

The Charter Analysis

Infringement of s. 2(b)

[29] There is no dispute that s. 517 of the Criminal Code infringes s. 2(b) of the Charter. That section provides as follows:

2. Everyone has the following fundamental freedoms: . . . . . b) freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication [page97 ]

[30] Notwithstanding the concession by the parties that s. 517 infringes s. 2(b), it is important to characterize the nature of that infringement. The ban on publication of bail proceedings is only temporary and expires if the accused is discharged after a preliminary inquiry or at the end of the trial. Further, the ban does not prevent anyone from attending in court to observe the proceedings.

[31] On the other hand, the ban is mandatory when sought by the accused and all-embracing; it prohibits publication of "the evidence taken, the information given or the representations made and the reasons, if any, given or to be given by the justice". Thus, even the reasons for releasing or detaining the accused, which may be of considerable interest to the public, cannot be published at the time the decision is made.

[32] The section therefore represents a dramatic curb on freedom of expression. In Canadian Broadcasting Corp. v. New Brunswick (Attorney General), 1996 184 (SCC), [1996] 3 S.C.R. 480, [1996] S.C.J. No. 38, at para. 18, La Forest J. wrote that the "freedom of individuals to discuss information about the institutions of government, their policies and practices, is crucial to any notion of democratic rule". He also observed, at para. 20, that "the courts, and particularly the criminal courts, play a critical role in any democracy". Section 517 cuts off meaningful and informed public debate about a fundamental aspect of the administration of criminal justice, the bail system, at the very time that the debate may be most important -- when the decision is made to grant or deny bail. It also hinders debate in other circumstances of great public interest, as where an accused on bail commits another, perhaps serious crime. The public is left to speculate about why the accused was released and the justice system is unable to provide a timely and meaningful response because of the statutorily imposed silence.

[33] While s. 517 does not prevent anyone from attending court to witness the operation of the bail system first hand, the section effectively prevents access to the courts for most of the population. As has been repeatedly said, very few people have the time to attend court proceedings; everyone else must depend upon the media for information about the court system. Section 517 thus interferes with the effective operation of the principle of open courts. As La Forest J. observed in Canadian Broadcasting Corp., at para. 23:

Openness permits public access to information about the courts, which in turn permits the public to discuss and put forward opinions and criticisms of court practices and proceedings. While the freedom to express ideas and opinions about the operation of the courts is clearly within the ambit of the freedom guaranteed by s. 2(b), so too is the right of members of the public to obtain information about the courts in the first place. [page98 ]

[34] Iacobucci and Arbour JJ. made similar observations in Vancouver Sun (Re), 2004 SCC 43, [2004] 2 S.C.R. 332, [2004] S.C.J. No. 41, at para. 26:

The open court principle is inextricably linked to the freedom of expression protected by s. 2(b) of the Charter and advances the core values therein: Canadian Broadcasting Corp. v. New Brunswick (Attorney General), supra, at para. 17. The freedom of the press to report on judicial proceedings is a core value. Equally, the right of the public to receive information is also protected by the constitutional guarantee of freedom of expression: Ford v. Quebec (Attorney General), 1988 19 (SCC), [1988] 2 S.C.R. 712; Edmonton Journal, supra, at pp. 1339-40. The press plays a vital role in being the conduit through which the public receives that information regarding the operation of public institutions: Edmonton Journal, at pp. 1339-40. Consequently, the open court principle, to put it mildly, is not to be lightly interfered with.

[35] Section 517 infringes the freedom of the press to comment on the courts and interferes with the "further freedom of the members of the public to develop and to put forward informed opinions about the courts" (Canadian Broadcasting Corp., at para. 26).

Section 1 analysis (1) Pressing and substantial objective

[36] The parties disagree about the objective of the legislation. The appellants submit that the objective of the legislation is to protect fair trial interests, the objective identified in Global Communications and by the application judge in his first decision. Canada and Ontario agree that this is an objective of the legislation, but also argue that the mandatory ban is meant to ensure expeditious show cause hearings. They argue, for example, that requiring the justice to embark on a hearing to consider whether a discretionary ban is appropriate after giving notice to the media could result in unfair and unreasonable delay in the bail proceedings.

[37] The appellants submit that this additional purpose could not have been contemplated in 1976, when the predecessor to s. 517 was enacted, since at that time, long before Dagenais was decided, there was no requirement to notify the media when a ban on publication was contemplated. They argue that the court ought not to consider this additional objective since, in accordance with R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, [1985] S.C.J. No. 17, 18 C.C.C. (3d) 385, it is not open to the Crown to rely on a shifting purpose that did not motivate those who enacted the legislation.

[38] In my view, the Crown's position does not attract the prohibition against shifting purpose. R. v. Butler, 1992 124 (SCC), [1992] 1 S.C.R. 452, [1992] S.C.J. No. 15, 70 C.C.C. (3d) 129 makes it clear that [page99 ]the court is entitled to consider the purpose of the legislation in the broadest sense. The interest in a fair trial embraces not simply the narrow interest of preventing potential jurors from being influenced by prejudicial material that might be disclosed at a bail hearing, but other interests intended to safeguard the accused's and society's interest in a fair trial. Those interests include preventing diversion of the accused's scarce resources to fight opposition to a publication ban and preventing delay of the bail hearing. As regards the latter, keeping accused in custody interferes with their ability to defend the case. The objectives of ensuring expeditious bail hearings, avoiding unnecessary detention of accused and allowing accused to retain scarce resources to defend their cases are all inextricably linked to the objective of ensuring a fair trial. To identify the objectives in this way does not require resort to the shifting purpose doctrine: see Butler, at pp. 493-94 S.C.R., p. 157 C.C.C.

[39] As Peter Hogg points out in Constitutional Law of Canada, 5th ed., looseleaf (Toronto: Carswell, 2007), at s. 38.9(a): "The statement of the objective should . . . be related to the infringement of the Charter, rather than to other goals. In other words, the statement of the objective should supply a reason for infringing the Charter right." In this case, the infringement is the mandatory delay in publication of bail proceedings, including evidence, submissions and reasons, upon application by the accused. I would characterize the objective of this infringement as follows:

Section 517 ensures that accused have a fair trial by an impartial jury, by providing the accused with an effective and expeditious means to prevent jurors from being exposed to prejudicial information that may be disclosed at the bail hearing.

[40] This objective is pressing and substantial. As pointed out in Dagenais, at p. 879 S.C.R., the accused's right to a fair trial is of such importance that it has been entrenched in ss. 7 and 11(d) of the Charter. The public also has an interest in a fair trial and the courts have "an interest in ensuring that justice [is] done, and an interest in safeguarding the repute of the administration of justice by ensuring that justice [is] seen to be done". The mandatory publication ban in s. 517 also helps protect the accused's liberty interest by allowing for an expeditious resolution of the bail issue. As I will discuss later, a discretionary ban imposed under s. 517 on application of the Crown or at common law at the request of the accused or the Crown could have other objectives. The issue at this point is the constitutionality of the mandatory ban, and that, in my view, can only be justified on fair trial grounds. [page100] (2) The proportionality test

[41] Where, as here, the government has established that the objective of the infringing measure is pressing and substantial, the court must consider whether the objective is proportional to the infringement. In this case, proportionality turns on the mandatory nature and scope of the publication ban. (i) Rational connection

[42] The first part of the proportionality test asks whether the measures adopted are rationally connected to the objectives. The measures must be carefully designed to achieve the objectives and must not be arbitrary, unfair or based on irrational considerations.

[43] In my view, the rational connection test has been met in this case. In the words of McLachlin J. in RJR-MacDonald Inc. v. Canada (Attorney General), 1995 64 (SCC), [1995] 3 S.C.R. 199, [1995] S.C.J. No. 68, at para. 153, the government must show "a causal connection between the infringement and the benefit sought on the basis of reason or logic". This causal relationship may be established by expert evidence, but such evidence is not always required, especially where the legislation is directed at changing human behaviour (RJR-MacDonald, at para. 154). As McLachlin J. noted, at para. 137 of RJR-MacDonald:

Discharge of the civil standard does not require scientific demonstration; the balance of probabilities may be established by the application of common sense to what is known, even though what is known may be deficient from a scientific point of view. (Citation omitted)

[44] The publication ban under consideration must be assessed in accordance with this standard. While there have been studies attempting to measure the impact of pre-trial publicity and the effectiveness of remedies such as jury instructions, the usefulness of this kind of evidence is open to debate, an issue I will return to below. As was said by Bastarache J. in Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, [2004] S.C.J. No. 28, at para. 77"Where the court is faced with inconclusive or competing social science evidence relating the harm to the legislature's measures, the court may rely on a reasoned apprehension of that harm." In my view, this is the type of case where it is open to the Crown to establish the rational connection without expert evidence.

[45] To understand why the rational connection test is met in this case, it is necessary to consider the purpose and nature of the [page101] bail hearing. As Bastarache J. stated in Thomson Newspapers Co. (c.o.b. Globe and Mail) v. Canada (Attorney General) (1998), 1998 829 (SCC), 38 O.R. (3d) 735, [1998] 1 S.C.R. 877, [1998] S.C.J. No. 44, at paras. 87-88, proportionality of the measures can only be evaluated by paying "close attention to detail and factual setting" and characterizing "the context of the impugned provision": see, also, R. v. Bryan, 2007 SCC 12, [2007] 1 S.C.R. 527, [2007] S.C.J. No. 12, 217 C.C.C. (3d) 97, at paras. 10-11.

[46] Part XVI of the Criminal Code deals with bail hearings. It sets out the rules for vindicating the accused's right to release pending trial where appropriate, which finds constitutional protection in s. 11(e) of the Charter, which guarantees an accused the right not to be denied reasonable bail without just cause. There are three fundamental aspects of the bail hearing that provide the context for considering the application of the proportionality test, and especially the rational connection aspect of that test.

[47] The first contextual factor is the scope of just cause for refusing to release the accused unconditionally. Section 515(10) defines just cause as the need (a) to ensure the accused's attendance at court; (b) to protect the safety of the public; or (c) to maintain confidence in the administration of justice. Section 515(10) thus sets the parameters for relevancy of information and evidence that may be adduced at the bail hearing. Evidence and information that would not be relevant at the eventual trial of the accused may be relevant at the bail hearing. For example, the fact that the accused was awaiting trial when he was alleged to have committed the offences for which bail is sought is highly relevant to all three aspects of just cause. But, it would only be in highly exceptional circumstances that this potentially prejudicial information would find its way into a trial. Similarly, the accused's criminal record, especially a record for similar offences, will be relevant to just cause, but the record will be adduced at the trial only in limited and closely defined circumstances. [See Note 3 below] Other highly prejudicial evidence, such as gang affiliation, might well not be admissible at trial but may be relevant to just cause and may be adduced at the bail hearing. Similarly, since the protection and safety of the public, including the alleged victim of the offence, is an important consideration at the bail hearing, evidence of the accused's relationship with the victim could be adduced at the bail hearing, despite the fact that the evidence may be prejudicial and not admissible at trial. [page102]

[48] The second contextual factor is the nature of the evidence that can be adduced at the bail hearing. Section 518 of the Criminal Code sets out the rules of admissibility at the bail hearing. Broadly speaking, the court may receive all evidence considered credible or trustworthy. As a result, the evidence is not put through the filter of the normal evidence rules that would apply at trial. Thus, for example, the prosecutor may introduce an accused's confession or wiretap evidence without having to establish that the evidence was lawfully obtained: see the discussion of the procedure and evidence at the bail hearing in chapter 5 of Trotter J.'s text.

[49] The third contextual factor is the timing of the bail hearing in the criminal process. Of necessity, the bail hearing is ordinarily held at the very earliest stages of the proceedings. Accused have limited information about the allegations against them. They often do not have retained counsel and instead rely upon duty counsel. They are not in a position to provide full instructions, even to retained counsel, about the allegations. In short, they are in a highly vulnerable position and have limited resources available to respond to the allegations against them. Thus, they may not be in a position to contradict or explain highly damaging but false allegations that the prosecutor makes at the bail hearing. The accused's vulnerability at this stage of the proceedings is an important contextual factor: Thomson Newspapers Co., at para. 90.

[50] With these contextual factors in mind, I return to the objective of the infringing measure: ensuring that the accused has effective and expeditious means to prevent jurors from obtaining prejudicial information that could preclude a fair trial. There is a logical causal relationship between this objective and the mandatory publication ban. Section 517 prevents information that could be highly prejudicial and inadmissible at trial from being disseminated to the public, some of whom will ultimately make up the jury in the accused's case.

[51] The mandatory nature of the prohibition enhances its effectiveness for this purpose. At the early stage of the bail hearing, the accused is not in a position to make effective submissions in favour of a discretionary ban. Accused have only a limited understanding of the case against them; their counsel, often duty counsel, will not be in a position to make informed submissions about whether evidence such as the accused's confession, criminal record or history with the victim will be admissible at the trial. Counsel cannot know with certainty whether the accused will be tried by a jury. Even where there is no possibility of a jury trial, as in summary conviction proceedings, the proceedings may [page103] be a precursor to more serious indictable proceedings. Neither can counsel know whether the trial will be relatively soon, while information about the bail hearing is fresh in the potential jurors' minds.

[52] The mandatory ban also serves the important interest of ensuring an expeditious bail hearing. If the accused were required to make submissions in favour of a discretionary ban, a judge might well need to adjourn the hearing so the accused may properly instruct counsel and obtain further information. Adjournments for this purpose would frustrate the objective of ensuring the accused have access to an expeditious method of protecting their fair trial rights.

[53] The media fairly make the point that there is limited empirical evidence to demonstrate the impact of pre-trial publicity of any kind, let alone bail proceedings, on jurors' ability to give the accused a fair trial. They point to the presumption that jurors will obey their oaths and will try the accused based only on the evidence admitted in court in accordance with the trial judge's instructions. Cory J. explained this presumption in his concurring reasons in Phillips v. Nova Scotia (Commission of Inquiry into the Westray Mine Tragedy), 1995 86 (SCC), [1995] 2 S.C.R. 97, [1995] S.C.J. No. 36. It was argued in that case that a commission of inquiry into a highly publicized event should be stayed until after the related criminal trial. Cory J. stated, at para. 133:

I am of the view that this objective [of a fair trial] is readily attainable in the vast majority of criminal trials even in the face of a great deal of publicity. The jury system is a cornerstone of our democratic society. The presence of a jury has for centuries been the hallmark of a fair trial. I cannot accept the contention that increasing mass media attention to a particular case has made this vital institution either obsolete or unworkable. There is no doubt that extensive publicity can prompt discussion, speculation, and the formation of preliminary opinions in the minds of potential jurors. However, the strength of the jury has always been the faith accorded to the good will and good sense of the individual jurors in any given case. (Emphasis added)

[54] However, the fact that a fair trial is obtainable in the vast majority of criminal trials, even in the face of considerable pre-trial publicity, does not, in my view, undermine the rational connection. Stating the proposition in this manner concedes that there may be some cases where a fair trial will not be possible despite jury instructions and the other measures that may be employed, such as adjournment of the trial, challenge for cause and change of venue. Cory J. acknowledged this in Phillips, pointing out, at para. 134, that one risk the Nova Scotia government took in proceeding with the inquiry while criminal charges [page104] were outstanding was that the charges would be stayed if a fair trial could not be achieved. At the bail hearing stage, it may be difficult to identify with reasonable certainty those cases where fair trial interests will be undermined by publication of the bail proceedings: see R. v. Daly, 2003 BCSC 1143, [2003] B.C.J. No. 1742, 178 C.C.C. (3d) 31 (S.C.), at para. 69, affd 2005 BCCA 389, [2005] B.C.J. No. 1697, 198 C.C.C. (3d) 185 (C.A.). A mandatory ban is therefore rationally connected to the aim of preventing juror tainting at a stage where it is difficult to assess the risk of an unfair trial based on prejudicial pre-trial publicity.

[55] I wish to briefly discuss the empirical information available concerning the impact of pre-trial publicity. I will revisit this evidence later, when considering the actual salutary effects of the legislation. As is well known, Dickson C.J.C. was sceptical about the value of this research seeking to demonstrate that juries are unable to properly disregard potentially prejudicial information. As he said in R. v. Corbett, 1988 80 (SCC), [1988] 1 S.C.R. 670, [1988] S.C.J. No. 40, at pp. 693-94 S.C.R.:

We should maintain our strong faith in juries which have, in the words of Sir William Holdsworth"for some hundreds of years been constantly bringing the rules of law to the touchstone of contemporary common sense" (Holdsworth, A History of English Law (7th ed. 1956), vol. I, at p. 349).

[56] For similar reasons, Barry J. in R. v. Kenny, 1991 2738 (NL SC), [1991] N.J. No. 253, 68 C.C.C. (3d) 36 (T.D.), at pp. 49-60 C.C.C., was of the view that the empirical research was of limited value. However, at p. 197 of his text, Trotter J. points to more recent studies than those referred to in Corbett and Kenny as suggesting that the negative impact of pre-trial publicity may be more profound than previously suspected. He writes as follows:

However, social science research suggests that our assumptions about how jurors are affected by adverse information relating to the accused person may be misplaced. A growing body of empirical literature supports the view that jurors are easily biased by learning of prejudicial information about the accused person. Moreover, instructions to disregard this prejudicial information are shown to be largely inefficacious. It may be necessary to reconsider the current orthodoxy relating to jury instructions. (Footnotes omitted)

[57] In my view, it is difficult to draw firm conclusions on the impact of pre-trial publicity from the empirical material. I will elaborate on this issue further below; at this point, it is sufficient for me to conclude that this is a case where the government should be entitled to rely upon reason or logic. In short, it is reasonable to believe that, in some cases, disclosure of material from the bail hearing could negatively affect potential jurors. As Cory J. said in Phillips, [page105] at para. 125"Perhaps science will one day be able to prove that in certain situations juror prejudice is inescapable. Until that time, common sense must provide guidance in these decisions." (ii) Minimal impairment

[58] The second element of the proportionality test is the minimal impairment analysis. The test at this stage is whether the mandatory prohibition in s. 517 abridges freedom of expression as little as is reasonably possible or, put another way, whether "the law falls within a range of reasonable alternatives": see RJR-MacDonald, at para. 160. So framed, the test necessarily requires the court to pay deference to choices made by Parliament even when imposing prior restraints on freedom of expression. That said, the Supreme Court of Canada has emphasized the importance of minimizing interference with information about the functioning of the courts. As Cory J. said in Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326, [1989] S.C.J. No. 124, at p. 1347 S.C.R.:

The importance of freedom of expression and of public access to the courts through the press reports of the evidence, arguments and the conduct of judges and judicial officers is of such paramount importance that any interference with it must be of a minimal nature.

[59] The appellants submit that s. 517 does not impair freedom of expression as little as is reasonably possible because there are alternatives that would be equally effective and trench less on freedom of expression. As Lamer C.J.C. said in Dagenais, at p. 879 S.C.R., the Charter fair trial guarantee "does not require that all conceivable steps be taken to remove even the most speculative risks".

[60] The media's principle submission on minimal impairment is that a discretionary ban that meets the test in Dagenais would adequately meet the fair trial objectives of s. 517. As noted in Dagenais, at p. 878 S.C.R., a discretionary ban may be ordered when: (a) Such a ban is necessary in order to prevent a real and substantial risk to the fairness of the trial, because reasonably available alternative measures will not prevent the risk; and (b) The salutary effects of the publication ban outweigh the deleterious effects to the free expression of those affected by the ban. (Emphasis in original)

[61] Canada and Ontario argue that a discretionary ban would not adequately protect the legislative objectives of s. 517. Canada particularly focuses on the delay inherent in a discretionary ban. [page106] An accused seeking a discretionary ban would have to adjourn the hearing to notify the media and prepare material to justify the order sought. Canada and Ontario also submit that, because of the early stage of the proceedings, the accused is not in a position to make informed submissions that could meet the Dagenais test. Similarly, because of the early stage of the proceedings and the lack of information, the presiding justice asked to impose the prohibition would not be in a position to make an informed decision.

[62] I agree with Ontario and Canada. This is a case where Parliament is mediating between the claims of competing groups -- namely, the claim of accused to a fair trial by an impartial jury uncontaminated by prejudicial and inadmissible evidence, and the claim of the press and the public to information. To paraphrase the words of the majority of the court in Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927, [1989] S.C.J. No. 36, at p. 994 S.C.R., the question is whether Parliament "had a reasonable basis, on the evidence tendered, for concluding that the [temporary mandatory ban on publication] impaired freedom of expression as little as possible given the government's pressing and substantial objective". In my view, there is a reasonable basis to conclude that a mandatory ban impairs freedom of expression as little as reasonably possible in order to meet the objective of ensuring that the accused has an effective and expeditious means to prevent jurors from obtaining prejudicial information that could be disclosed at a bail hearing but would preclude a fair trial.

[63] I will review a number of less impairing alternatives to show why these alternatives would not adequately achieve the law's objective. First, the media suggest that a less rights- impairing solution is to allow for a time limited publication ban, pending a hearing to determine if the ban should continue. In the meantime, the bail hearing would proceed. It seems to me, however, that this proposed solution ignores the contextual factors discussed above and places the accused in an invidious position. Without knowing whether or not the publication ban will be in place, the accused cannot know whether to take the risk of contesting bail to possibly obtain the immediate reward of release at the cost of the more serious risk of poisoning the minds of jurors at the subsequent trial. Either accused wait until the publication ban issue has been finally determined, or they forego the bail application until they are in a better position to make reasoned submissions in accordance with the Dagenais test. This would have a detrimental effect on an accused's liberty interest. As Martin L. Friedland said in [page107] Detention Before Trial: A Study of Criminal Cases Tried in the Toronto Magistrates' Court (Toronto: University of Toronto Press, 1965), at p. 172"The law should abhor any unnecessary deprivation of liberty and positive steps should be taken to ensure that detention before trial is kept to a minimum" (footnote omitted).

[64] The "certainty" argument in favour of mandatory prohibitions in bail hearings is similar to the rationale used to uphold the mandatory ban in sexual assault cases in Canadian Newspapers Co. v. Canada (Attorney General), 1988 52 (SCC), [1988] 2 S.C.R. 122, [1988] S.C.J. No. 67. In that case, the court held that only the certainty provided by a mandatory ban preventing publication of the identity of the complainant could achieve Parliament's objective of encouraging the reporting of sexual offences. As the court said, at p. 132 S.C.R.:

Obviously, since fear of publication is one of the factors that influences the reporting of sexual assault, certainty with respect to non-publication at the time of deciding whether to report plays a vital role in that decision. Therefore, a discretionary provision under which the judge retains the power to decide whether to grant or refuse the ban on publication would be counterproductive, since it would deprive the victim of that certainty. Assuming that there would be a lesser impairment of freedom of the press if the impugned provision were limited to a discretionary power, it is clear, in my view, that such a measure would not, however, achieve Parliament's objective, but rather defeats it. (Emphasis in original)

[65] Accused may also be placed in a position that deprives them of other constitutional rights. For example, they may have to seek to adjourn their trial until the sting of publicity has faded, thus foregoing their s. 11(b) rights to a trial within a reasonable time. Alternatively, accused may have to waive their s. 11(f) right to a jury trial in order to obtain a trial within a reasonable time.

[66] A second possible means of impairing the right as little as possible would be to permit publication of the bail proceedings at the time but impose a discretionary ban closer to the trial. This is a serious alternative to the mandatory ban effective at the time of the bail hearing, and it is not without some doubt that I hold that the government has nevertheless met the minimal impairment test. The alternative of a future publication ban does not meet the problem that accused need some certainty in order to decide how to conduct the bail hearing. The notion of a future publication ban also does not meet the problem of intense media attention given to the information from the bail hearing when the trial will occur soon after the bail hearing, perhaps because the accused ends up being detained in custody. [page108]

[67] Further, as Hogg cautions at s. 38.11(b), courts must be aware of the "practicalities of designing and administering a regulatory regime" and cannot be "indifferent to considerations of cost". Requiring an accused to fight for a discretionary ban at any time involves expenditure of scarce resources that could be better used to defend the case.

[68] A final alternative would be to impose a limited discretionary ban on only prejudicial evidence, such as the accused's confession and other presumptively inadmissible evidence. The concern here is that at the bail hearing stage, where there is only limited disclosure, the accused has had only a limited opportunity to consult counsel, and the rules of admissibility are relaxed, it will be difficult for the accused and counsel, usually duty counsel, to know what information will be prejudicial. To require counsel and the justice of the peace to sort through the information at this early stage could result in delay. Further, such a ban would still not solve the difficulties of coping with intervention by the media; the potentially prejudicial material may be the very information the media seek to disclose to the public. In fairness, it is probably this information that the public requires to understand the bail decision.

[69] That said, I will consider these alternatives again when I deal with the final element of the proportionality test, which requires the court to measure the salutary and deleterious effects of the mandatory ban. The fact that there exist alternatives to the mandatory ban that would have a less detrimental effect on freedom of expression is an important consideration when measuring the salutary effects of the measure. (iii) Proportionate effect

[70] The final element of the proportionality test requires that the court be satisfied that the effects of the limiting measure are proportionate to the objectives, and that the actual salutary effects of the law outweigh its deleterious effects. In my view, it is here that s. 517 fails. In Dagenais, Lamer C.J.C., although considering the test for imposition of a common-law publication ban, discussed at length this third element of the proportionality test under s. 1. He made the following points, at pp. 887-89 S.C.R.:

In many instances, the imposition of a measure will result in the full, or nearly full, realization of the legislative objective. In these situations, the third step of the proportionality test calls for an examination of the balance that has been struck between the objective in question and the deleterious effects on constitutionally protected rights arising from the means that have been employed to achieve this objective. At other times, however, the measure at issue, while rationally connected to an important objective, will [page109] result in only the partial achievement of this object. In such cases, I believe that the third step of the second branch of the Oakes test requires both that the underlying objective of a measure and the salutary effects that actually result from its implementation be proportional to the deleterious effects the measure has on fundamental rights and freedoms . . . [E]ven if the importance of the objective itself (when viewed in the abstract) outweighs the deleterious effects on protected rights, it is still possible that the actual salutary effects of the legislation will not be sufficient to justify these negative effects. . . . . .

[I]n appropriate cases it is necessary to measure the actual salutary effects of impugned legislation against its deleterious effects, rather than merely considering the proportionality of the objective itself. . . . . .

I believe that even if an objective is of sufficient importance, the first two elements of the proportionality test are satisfied, and the deleterious effects are proportional to the objectives, it is still possible that, because of a lack of proportionality between the deleterious effects and the salutary effects, a measure will not be reasonable and demonstrably justified in a free and democratic society. (Underlining in original; italics added)

[71] In the result, Lamer C.J.C. rephrased this last part of the proportionality test as follows, at p. 889 S.C.R.:

. . . there must be a proportionality between the deleterious effects of the measures which are responsible for limiting the rights or freedoms in question and the objective, and there must be a proportionality between the deleterious and the salutary effects of the measures. (Emphasis in original)

[72] Or, as it was put by Hogg at s. 38.12, this third element of the proportionality tests requires "a balancing of the objective sought by the law against the infringement of the Charter. It asks whether the Charter infringement is too high a price to pay for the benefit of the law."

[73] To deal with the first part of the test: is there proportionality between the deleterious effects of the mandatory publication ban and the objective? Given that the ban is only temporary and that the court proceedings themselves are open to the press and the public, I think it is arguable that the effects of the mandatory ban are proportionate to the objectives. As Lamer C.J.C. pointed out in Dagenais, at p. 890 S.C.R., when considering this element of the proportionality test it is important to bear in mind that the pressing and substantial objective at issue, preserving the accused's fair trial right, is itself a fundamental right and "as such, it is a matter of exceptional importance". [page110]

[74] Where the mandatory ban fails the s. 1 test is the requirement of proportionality between the deleterious and the salutary effects of the measure. In my view, the deleterious effects of s. 517 are substantial and the salutary effects are in many respects speculative and limited. (a) Deleterious effects

[75] What then are the deleterious effects of the mandatory publication ban in s. 517 on freedom of expression? First, the ban prohibits publication of all information, evidence, and submissions, even those that could not possibly affect the accused's fair trial. The ban also prohibits publication of this material where alternative measures, such as challenge for cause, change of venue, and strong judicial directions could eliminate risk to a fair trial. All that can be published is the fact of the bail hearing and the result: see Daly, at para. 63.

[76] Second, the ban prohibits the publication of the reasons of the justice or judge for releasing or detaining the accused, even reasons that are carefully measured to explain the decision but not interfere with the accused's fair trial rights. This is, in my view, a serious infringement of freedom of expression, especially in high-visibility cases. The public is left to wonder, for example, why in light of the serious allegations made against the accused at the pre-hearing press conference a justice or judge nevertheless released the accused. As Berger J.A. said in R. v. White, 2005 ABCA 435, [2005] A.J. No. 1727, 376 A.R. 63 (C.A.) ("White #1"), at para. 16"How is public confidence [in the administration of justice] enhanced by a 'cone of silence' descending over the careful and considered analysis of the adjudicator?"

[77] Nordheimer J. put the matter very well in R. v. Smith, 2006 30593, at para. 2:

[O]ne unfortunate consequence of the mandatory order banning the evidence heard and reasons given in a bail application is that the public is denied the very information that they ought to have to properly understand and assess any order the court may make and the reasons underlying that order. That denial of information has the very real potential for fostering misunderstanding, and a lack of proper appreciation among members of the public respecting the result, and the specific facts of the case that must of necessity, solely dictate that result. (Emphasis added)

[78] In R. v. Hall, 2002 SCC 64, [2002] 3 S.C.R. 309, [2002] S.C.J. No. 65, 167 C.C.C. (3d) 449, McLachlin C.J.C. wrote, at para. 27"Public confidence is essential to the proper functioning of the bail system and the justice system as a whole." Again, at para. 31, she wrote: "Without public confidence, the bail system and the justice system [page111] generally stand compromised." Section 517, however, prevents the dissemination of the information necessary to promote public confidence in the bail system. This is the case both at the time the initial decision is made, and later where subsequent events may raise questions as to the wisdom of that decision, as when the accused who is on bail commits further offences. The fact that an accused was on bail when he or she committed a further serious offence often receives wide coverage in the media. However, the public is left to speculate as to why the accused was initially released, because of the s. 517 order.

[79] Cory J. touched on this aspect of publication of court proceedings in Edmonton Journal in considering provincial legislation banning publication of details relating to matrimonial proceedings, at pp. 1341-42 S.C.R.:

[T]he comments of counsel and the presiding judge are excluded from publication. How then is the community to know if judges conduct themselves properly. How will it know whether remarks might have been made, for example, that a wife should submit to acts of violence from her husband or that a wife should endure the verbal abuse or blows of her husband. The community has a right to know if such remarks are made yet if there is no right to publish, the judge's comments may be hidden from public view. Thus it can be seen that the effect of s. 30(1) is to repress the publication of important aspects of court proceedings. The prohibitions are unnecessarily extensive. (Emphasis added)

[80] Third, the ban applies in cases that could never be tried by a jury and also in cases where there is no likelihood of the case ever being tried by a jury. Less than 2 per cent of all criminal cases are tried by jury. [See Note 4 below] Thus, in the overwhelming majority of cases in which a publication ban is imposed and freedom of expression infringed, the ban does not serve the objective of protecting the accused from a jury tainted by pre-trial publicity. A ban on publication may serve other objectives in some cases, an issue to which I will return below, especially when considering the recent decision of the Alberta Court of Appeal in R. v. White, 2008 ABCA 294, [2008] A.J. No. 956, 93 Alta. L.R. (4th) 239 (C.A.) ("White #2") and the appropriate remedy in this case. However, the issue here is the constitutionality of the mandatory ban on publication and the objective of the mandatory ban is protecting the fair trial right of the accused. [page112]

[81] Fourth, freedom of expression is infringed in cases such as this where there has already been considerable pre-hearing publicity and there is no indication that what would be disclosed at the bail hearing would be any more prejudicial than information disclosed by the authorities at the time of the arrest. It is of interest in this case that some of the accused opposed the publication ban precisely because they hoped that the more measured information disclosed at the bail hearings would counteract the inflammatory information disclosed by the police.

[82] Fifth, the ban extends to the time of discharge at the preliminary inquiry or the conclusion of the trial, which can be months if not years away. Ontario and Canada argue that the ban is only temporary and that this moderates the deleterious effect on freedom of expression. This ignores the fact that the time when the public is most interested in the information disclosed at the bail hearing is when the bail decision is made. That is the point in the proceedings when the public is entitled to scrutinize and hold the criminal justice system to account. As Doherty J.A. said, at para. 40 of R. v. Domm (1996), 1996 1331 (ON CA), 31 O.R. (3d) 540, [1996] O.J. No. 4300, 111 C.C.C. (3d) 449 (C.A.), leave to appeal to S.C.C. refused [1997] S.C.C.A. No. 78, 43 C.R.R. (2d) 188:

The values promoted by s. 2(b) are not served by publication when the speaker has lost his audience and the message to be conveyed has lost its purpose.

[83] The ban on publication stifles informed public participation in one of the crucial aspects of the criminal justice system. As Cory J. said in Edmonton Journal, at p. 1340 S.C.R."Discussion of court cases and constructive criticism of court proceedings is dependent upon the receipt by the public of information as to what transpired in court. Practically speaking, this information can only be obtained from the newspapers or other media": also see, Dagenais, at p. 883 S.C.R.

[84] Sixth, in the case of jointly charged accused, as in this case, the publication ban infringes the rights of those co-accused who did not wish the ban. Such a ban infringes the interests of those accused to a public hearing. While the constitutional right to a public hearing as guaranteed by s. 11(d) of the Charter likely has no direct application at the bail hearing stage, [See Note 5 below] an accused has an interest in openness of the court process, which includes media access to the courtroom and reporting on the proceedings: [page113] R. v. Mentuck, 2001 SCC 76, [2001] 3 S.C.R. 442, [2001] S.C.J. No. 73, 158 C.C.C. (3d) 449, at paras. 30 and 52.

[85] Seventh, as a result of the ban on publication of proceedings at the bail hearing, individuals with helpful information might not learn about the case. Such information might be important to the defence or the prosecution, but might never come to light or may not come to light until much later in the proceedings, perhaps at the trial, leading to disruption of the proceedings when the parties must attempt to investigate the new information: see Dagenais, at p. 883 S.C.R. (b) Salutary effects

[86] As to the salutary effects of the legislation, in my view they are generally limited and largely speculative. First, the paramount objective of the publication ban is to protect against juror contamination. In my discussion of rational connection, I have accepted that, as a matter of logic and reason, there is a causal connection between this objective and the publication ban. It is now time to examine the strength of that causal connection in attempting to measure the actual salutary effects of the mandatory ban.

[87] Earlier, I referred to research discussed by Trotter J. in his text. It is difficult to draw firm conclusions from this material. As an example, one of the most ambitious studies, Amy Otto et al."The Biasing Impact of Pretrial Publicity on Juror Judgments" (1994), 18 Law and Human Behaviour 453, provided subjects with simulated press reports immediately before they viewed an edited videotape of an actual trial. The researchers concluded, at p. 464, that the results demonstrated that pre- trial publicity does have an impact, with the strongest effects being from negative pre-trial publicity about the accused's character. This study and other studies referenced by Trotter J. tend to deprecate the efficacy of jury instructions; some studies actually show that jury instructions can aggravate the prejudicial impact of certain kinds of information. [See Note 6 below] The difficulty with this study is that it fails to take into account the impact of delay, as bail hearings typically take place months if not years before trials.

[88] An earlier study by Geoffrey P. Kramer et al."Pretrial Publicity, Judicial Remedies, and Jury Bias" (1990), 14 Law and Human Behavior 409, found, at pp. 431-32, [page114] that even relatively short delays (in this study 12 days) "eliminated bias created by exposure to factually biasing publicity". The delay did not have the same impact on what the authors refer to as "emotionally biasing" information. Factually biasing publicity is publicity that contains information bearing on the perceived guilt of the accused. Emotionally biasing publicity is publicity that arouses the emotions of the public but has no direct evidentiary relevance. Examples of the former are reports of a confession, prior criminal record, or inadmissible but relevant evidence. This is the kind of material that is most likely to be raised at a bail hearing. Examples of emotionally biasing publicity would be graphic or lurid depictions of the murder victim's injuries or information that creates a climate of fear, such as publicity surrounding a series of child murders. [See Note 7 below] This latter type of information could be disclosed at the bail hearing, but it is also the very type of information that finds its way into the public realm in any event through sources not subject to the s. 517 order, such as police reports or interviews with neighbours.

[89] In the circumstances, it is difficult to quarrel with the observations in the Ontario Royal Commission Inquiry into Civil Rights, Report Number One, vol. 2 (Ottawa: Queen's Printer, 1968), at p. 763, made in relation to publication of proceedings at preliminary inquiries:

The risk is that wide publication of the evidence given at a preliminary inquiry may so influence the minds of prospective jurors that they will approach their duties with conscious or unconscious bias against the accused, or that the public mind in the community will be so aroused that members of a trial jury will be intimidated by the force of public opinion. We have grave doubts that this is a real risk. In the first place, at the time of a preliminary inquiry no one, including the jurors, knows who the jurors at the trial will be. In metropolitan areas the public memory is very short and individuals are largely anonymous. Few members of the public can remember what they have read or heard about a particular case for many days, let alone months. In less populous areas gossip and rumour spread more easily, but gossip and rumour thrive on secrecy. It is much more likely that vicious and inaccurate gossip and rumour will be spread throughout a community by individuals who claim to have knowledge than by a fair and accurate report by news media. (Emphasis added)

[90] Even granting the possibility that pre-trial publicity can influence jury behaviour, this would only be the case in the most unusual instances, being cases where a single powerful piece of inadmissible evidence comes to the attention of the jury pool at a [page115] time proximate enough to the trial to have an impact. For example, in Phillips, at para. 162, Cory J. suggested that rather than a stay of the public inquiry, an appropriate remedy would be a ban on publication of the accused's testimony at the inquiry to prevent potential jurors from being exposed to testimony that they might never hear at the trial. An accused's confession might fall within the same category: see Linden, Limitations on Media, at p. 320.

[91] In R. v. White, 2007 ABQB 359, [2007] A.J. No. 608, 221 C.C.C. (3d) 393 (Q.B.) (reversed by (White #2)), the media challenged the constitutionality of s. 517 and adduced evidence from a psychologist on the impact of the pre-trial publicity. It was the psychologists' opinion that pre-trial publicity is unlikely to negatively impact the impartiality of jurors except where a single piece of information that essentially decides the case is disclosed to jurors before trial but not adduced at trial: see the Court of Queen's Bench decision at para. 71 and the Court of Appeal decision at para. 40.

[92] In cases where a single piece of highly prejudicial but inadmissible evidence is led at a bail hearing, it should be possible to craft remedial measures to ensure that the accused will receive a fair trial; for example, a ban on publication of that piece of evidence, a publication ban proximate to the trial, challenge for cause or change of venue. An example of a remedial measure can be seen in R. v. Lake, [1997] O.J. No. 5446 (Gen. Div.), at para. 24. In that case, a witness sought a ban on publication of his testimony because he was to be tried for the same offence a year later. McCombs J. granted the order but only to take effect seven days after completion of the Lake trial. As he said, at para. 26:

In my view, the principle of freedom of the press to publish contemporaneous accounts of this trial outweighs the speculative risk to the fairness of [the witness's] trial which will not take place until at least a year from now. On the other hand, once the present trial has been completed, the analysis changes. After this trial is over, the right of the press to publish further accounts of it assumes much less importance, while at the same time, the risk of prejudice to the fairness of [the witness's] pending trial becomes substantially greater, particularly if there is continuing sustained publicity of [the witness's] compelled testimony.

[93] Commentary from judges in Canada and other common-law jurisdictions is virtually uniform that the impact of pre-trial publicity is speculative and that other measures short of a contemporaneous ban on publication of the entire proceedings will protect the accused's fair trial rights:

To begin, I doubt that jurors are always adversely influenced by publications. There is no data available on this issue. [page116] However, common sense dictates that in some cases jurors may be adversely affected. Assuming this, I nevertheless believe that jurors are capable of following instructions from trial judges and ignoring information not presented to them in the course of the criminal proceedings: Dagenais, at p. 884 S.C.R., per Lamer C.J.C.

[T]his Court [in R. v. Corbett] has strongly endorsed the ability of a jury to follow the explicit instructions of a judge. This endorsement surely applies as much to the instruction to ignore all information not presented in the course of the criminal proceedings as it applies to the instruction to use evidence of prior convictions for one purpose and not another: Dagenais, at p. 885 S.C.R., per Lamer C.J.C.

[T]he risk of prejudice to an accused's fair trial rights from pre-trial publicity is highly speculative: Phillips, at para. 34, per L'Heureux-Dubé J.

I cannot accept the contention that increasing mass media attention to a particular case has made this vital institution either obsolete or unworkable. There is no doubt that extensive publicity can prompt discussion, speculation, and the formation of preliminary opinions in the minds of potential jurors. However, the strength of the jury has always been the faith accorded to the good will and good sense of the individual jurors in any given case: Phillips, at para. 133, per Cory J.

There is a feature of our trial system which is sometimes overlooked or taken for granted. The collective experience of this constitution as well as the previous constitution of the court, both when we were in practice at the Bar and judicially, has demonstrated to us time and time again, that juries up and down the country have a passionate and profound belief in, and a commitment to, the right of a defendant to be given a fair trial. They know that it is integral to their responsibility. It is, when all is said and done, their birthright, it is shared by each one of them with the defendant. They guard it faithfully. The integrity of the jury is an essential feature of our trial process. Juries follow the directions which the judge will give them to focus exclusively on the evidence and to ignore anything they may have heard or read out of court. No doubt in this case Butterfield J. will give appropriate directions, tailor-made to the individual facts in the light of any trial post the sentencing hearing, after hearing submissions from counsel for the defendants. [page117] We cannot too strongly emphasise that the jury will follow them, not only because they will loyally abide by the directions of law which they will be given by the judge, but also because the directions themselves will appeal directly to their own instinctive and fundamental belief in the need for the trial process to be fair: R. v. B., [2006] E.W.C.A. Crim 2692, [2007] E.M.L.R. 145 (C.A. Crim. Div.), at para. 31, per Judge J.

But the entire system of trial by jury is based upon the assumption that the jury will follow the instructions which they receive from the trial judge and that they will return a true verdict in accordance with the evidence: Montgomery v. HM Advocate, [2003] 1 A.C. 641 (P.C.), at p. 674 A.C., per Lord Hope of Craighead.

[94] In its Working Paper 56, Public and Media Access to the Criminal Process (Ottawa, 1987), the Law Reform Commission recommended repeal of the mandatory publication ban in what is now s. 517, taking the view, at p. 76, that this provision and the similar provision banning publication of preliminary inquiry proceedings "are overly restrictive of freedom of expression" and "constitute an unjustifiable intrusion on the principle of maximum openness".

[95] A further point attenuating the salutary effect of the publication ban of material from the bail hearing is that such a ban has no effect on other pre-trial publicity, some of which may be more inflammatory than the generally factual information presented at the bail hearing. Bail hearings that the media are most likely to want to publicize are the very ones that are most likely to be attended by substantial pre-trial publicity, as in this case. A ban on publication of the bail hearing proceedings prevents re-publication of this information at a time proximate to the trial. However, such a ban has no impact on the media's ability to re-publish other material revealed at the time of the arrest or acquired outside the bail hearing in the months following the initial arrest.

[96] The second and most significant salutary effect of the mandatory publication ban is the avoidance of delay and expense for an accused who seeks the publication ban. Because the s. 517 ban is available simply upon request, the accused need not establish any basis for the order and the court need not take any steps to notify the media. If the mandatory ban in s. 517 is struck down, an accused seeking a discretionary ban would have to meet the test set out in Dagenais, at p. 878 S.C.R., or Mentuck, at para. 32, [page118] and show that the ban is necessary to prevent "a real and substantial risk to the fairness of the trial" or "a serious risk to the proper administration of justice". In Mentuck, the court also held, at para., 34 that "it must be a risk the reality of which is well- grounded in the evidence". The accused must show that reasonably available alternative measures will not prevent the risk and that the salutary effects of the ban outweigh the deleterious effects to freedom of expression. The Crown faces the same test when it seeks a discretionary ban under s. 517.

[97] I am not convinced that the delay and other problems suggested by the government and the accused are as serious as argued. First, as to notice, the justice has a discretion whether to give notice and the form of notice: Dagenais, at p. 869 S.C.R. Given the need for an expedited bail hearing to avoid the unwarranted detention of the accused, it would seem to me that in most cases the justice need not give notice to the media. Such an approach appears to be authorized by Named Person v. Vancouver Sun, 2007 SCC 43, [2007] 3 S.C.R. 253, [2007] S.C.J. No. 43, where the court held, at paras. 53-54, that notice need not be given every time an in camera proceeding is to take place:

The decision to post a public notice regarding the existence of the proceeding is a matter of discretion on the part of the judge. In other words, no one has a right, constitutional or otherwise, to be informed of all situations in which informer privilege is claimed. The reason for this is simply practical: there is no real difference -- vis-à-vis the open court principle -- between a situation in which informer privilege exists and any other situation in which some part of a proceeding takes place in camera -- be it a situation of a child sexual assault victim, or a situation involving solicitor-client privilege. It would be unworkable and unreasonable to expect that literally every time an in camera proceeding is taking place, a judge has the obligation to publicize its existence and invite submissions from all comers on whether that proceeding should be held in camera. Nor should a judge choose "worthy" interveners.

Instead, the judge retains discretion as to whether or not to provide public notice of the in camera proceeding involving informer privilege. The exercise of the discretion will depend on the circumstances, such as whether the holder of the privilege is present and plays an active role in court, for instance, as was the case here. Whether the judge issues notice, or (as can certainly happen) the media independently learns of the existence of the in camera proceeding, the next step in the procedure is to hear submissions to determine the extent of the need for in camera proceedings. It is at this point that the media is granted standing to present arguments on how informer privilege can be respected with minimal effect on the open court principle. (Emphasis added)

[98] Similarly, it would be unworkable and unreasonable to expect that every time an accused applies for a ban on publication at a show cause hearing, the justice of the peace has the obligation [page119] to halt the proceedings and somehow give notice to the media. Hundreds of show cause hearings take place in this country every day. Even if accused sought publication bans in a small percentage of those cases, it is simply not practicable to expect justices of the peace to give notice. It is important that bail hearings proceed expeditiously, and this will usually require that no notice of the accused's application is given.

[99] As to the issue of the accused needing time to gather "evidence" of the risk to trial fairness or the administration of justice, this issue must be put in context. As already noted, the strict rules of evidence do not apply at bail hearings. In my view, in considering whether to impose a publication ban, the justice of the peace is entitled to act upon the kind of information that can be received on a bail hearing, being evidence that is considered credible and trustworthy. It is not unusual for evidence to be presented through statements of facts read out by Crown counsel. It would seem to me that the justice of the peace would be entitled to act upon submissions of counsel in deciding whether or not to impose a publication ban. It is not practicable to expect the party seeking the order to adduce the kind of evidence that might be admissible at a trial, in accordance with the normal rules of evidence.

[100] The Crown argues that a third salutary effect of the publication ban is the prevention of witness contamination. The Supreme Court of Canada has rejected mere assertions that pre- trial publicity taints potential witnesses in the search warrant context. I can see no reason to believe that the situation is any different where the information is disclosed at the bail hearing. In Toronto Star Newspapers Ltd. v. Ontario, 2005 SCC 41, [2005] 2 S.C.R. 188, [2005] S.C.J. No. 41, at para. 36, the Crown relied upon an assertion from a police officer that disclosure of material in an information to obtain a search warrant could taint witnesses. Fish J., speaking for the court, at paras. 38-39, adopted this court's reasons for rejecting this broad assertion:

In Doherty J.A.'s view, if that general proposition were sufficient to obtain a sealing order,

. . . the presumptive rule would favour secrecy and not openness prior to trial. A general assertion that public disclosure may distract from the ability of the police to get at the truth by tainting a potential witness's statement is no more valid than the equally general and contrary assertion that public disclosure enhances the ability of the police to get at the truth by causing concerned citizens to come forward with valuable information.

[101] A fourth, and related, salutary effect is the protection of ongoing criminal investigations. The public incidentally benefits from the mandatory ban sought by an accused where there are [page120] ongoing investigations that could be compromised because of information disclosed at the bail hearing. Because the ban is mandatory at the insistence of the accused, the Crown need not seek its own ban and therefore need not publicly disclose that other investigations are extant. However, this issue of compromising ongoing investigations can be dealt with by a more limited ban on application by the Crown of only the material that could compromise the investigation: see Mentuck, at paras. 46-47. In appropriate circumstances, the application could be done in camera, as was done by this court in R. v. Y. (K.S.) (October 11, 2002), unreported (Ont. C.A.).

[102] A fifth incidental salutary effect of the mandatory ban is that character witnesses, employers and others who can assist an accused in establishing grounds for release may be more likely to come forward, especially for accused charged with high-profile or notorious crimes, if they know that their testimony would not be published. Similarly, potential sureties may be more likely to testify if they know their testimony can not be published. On the other hand, it is unlikely that the s. 517 prohibition covers the identity of sureties and there is no requirement that sureties always testify at the bail hearing. [See Note 8 below]

[103] Sixth, a publication ban also advances the privacy interests of the accused, witnesses and victims. While those privacy interests might not be sufficient on their own to justify interference with freedom of expression, protection of those privacy interests are nevertheless one of the salutary effects of the legislation. In other contexts, privacy interests have been held to justify some limitations on freedom of expression: see Edmonton Journal, at p. 1345 S.C.R. In some circumstances, protecting privacy interests by banning publication or limiting access to court proceedings may advance the search for truth, one of the objectives underlying freedom of expression itself: see Canadian Broadcasting Corp., at paras. 34 and 38-43, and Sierra Club of Canada v. Canada (Minister of Finance), 2002 SCC 41, [2002] 2 S.C.R. 522, [2002] S.C.J. No. 42, at paras. 70 and 77. Thus, a publication ban on bail proceedings could encourage accused to adduce evidence of a very private nature, such as evidence about mental health or addiction that might assist the court in reaching a correct decision on bail. That said, it is not at all apparent that a mandatory ban in all cases is necessary to protect the interests of those few accused. In those few cases counsel, by articulating the reason for a limited ban, would be able to meet the test for a discretionary ban. [page121]

[104] As was pointed out in Thomson Newspapers Co., at para. 91, a contextual factor to be taken into account is the nature of the activity that is infringed. In R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697, [1990] S.C.J. No. 131, 61 C.C.C. (3d) 1, Dickson C.J.C. upheld the criminal prohibition on wilful promotion of hatred. In considering this third component of the proportionality test, he took into account, at p. 787 S.C.R., p. 66 C.C.C., that hate speech is "only tenuously connected with the values underlying the guarantee of freedom of speech" and that the section was narrowly drawn. Section 517 by contrast is broadly drawn and infringes a crucial aspect of the democratic commitment. (iv) A comment on the efficacy of publication bans in the face of current technology

[105] The question of the efficacy of publication bans does not fit easily under the headings of either the salutory or deleterious effects. In Dagenais, at p. 886 S.C.R., Lamer C.J.C. framed the problem in this way:

It should also be noted that recent technological advances have brought with them considerable difficulties for those who seek to enforce bans. The efficacy of bans has been reduced by the growth of interprovincial and international television and radio broadcasts available through cable television, satellite dishes, and shortwave radios. It has also been reduced by the advent of information exchanges available through computer networks. In this global electronic age, meaningfully restricting the flow of information is becoming increasingly difficult. Therefore, the actual effect of bans on jury impartiality is substantially diminishing.

[106] On the one hand, the salutory effect of any publication ban is undermined by the ease with which the ban can be circumvented. On the other hand, because of the nature of the Internet, information first published at the time of the bail hearing is always accessible, right up to the time of the trial. In other words, the court cannot always simply rely upon the fact that time will have passed from when the information was first published and that this passage of time will lessen any prejudicial effects of the information.

[107] On the whole, perhaps all that can be said about the efficacy of publication bans in the era of mass communication and the Internet is that the salutory and deleterious effects are uncertain. (v) Conclusion on proportionality

[108] To conclude, in my view, the deleterious effect of the mandatory publication ban outweighs the salutary effects. I appreciate that others might weigh the salutory and deleterious factors [page122] differently and it is hardly an exact science. However, when one looks at the severity of the deleterious effects measured against the speculative nature of many of the claimed salutory effects, I am not convinced that the government has met this aspect of the s. 1 test.

[109] In saying this, I have not lost sight of the constitutional dimension to the mandatory ban. In R. v. Pearson, supra, Lamer C.J.C. noted that the presumption of innocence as guaranteed by s. 11(d) of the Charter is not confined to the trial requirement that the prosecution prove its case beyond a reasonable doubt. As he said, at p. 683 S.C.R., p. 136 C.C.C."the starting point for any proposed deprivation of life, liberty or security of the person of anyone charged with or suspected of an offence must be that the person is innocent". He also noted, at p. 688 S.C.R., p. 139 C.C.C."Section 11(e) entrenches the effect of the presumption of innocence at the bail stage of the criminal process."

[110] There has been little exploration of the procedural content to the right to reasonable bail in s. 11(e). In Pearson, at p. 690 S.C.R., p. 140 C.C.C., Lamer C.J.C. held that s. 11(e) "creates a broad right guaranteeing both the right to obtain bail and the right to have that bail set on reasonable terms". It would seem to me that an aspect of the right to obtain bail is the right to have an expeditious bail hearing under circumstances where the accused can adduce evidence necessary to promote the right to have bail set on reasonable terms. The mandatory ban promotes the right to an expeditious bail hearing. Because the accused need not demonstrate any basis for the order, there will be no delay while he or she assembles whatever information may be necessary to justify the order. The mandatory ban may also assist the accused in assembling witnesses and sureties and thus promote the right to have bail set on reasonable terms.

[111] However, because the basis for the publication ban can be established through submissions, rather than in accordance with the strict rules of evidence, and because notice need not be given to the media, I am not persuaded that there will be unreasonable delay where an accused seeks a publication ban. As to the problem of being unable to obtain witnesses or sureties without the assurance of the publication ban, while I do not doubt that this is a concern in some rare cases, it is not a concern that justifies a mandatory ban in all cases. If an accused in a particular case can make out a basis for a limited publication ban on those grounds, the appropriate remedy is to grant a discretionary ban.

[112] Thus, the part of s. 517 that authorizes a mandatory ban on the application of the accused fails the proportionality test. [page123]

Reconsidering Global Communications

[113] Canada and Ontario submit that this court should not overrule Global Communications. In David Polowin Real Estate Ltd. v. Dominion of Canada General Insurance Co. (2005), 2005 21093 (ON CA), 76 O.R. (3d) 161, [2005] O.J. No. 2436 (C.A.), leave to appeal refused [2005] S.C.C.A. No. 388, Laskin J.A. discussed at length the circumstances in which this court should overrule one of its own decisions. He referred to the five factors that the Supreme Court of Canada has articulated that would allow it to overrule one of its previous cases as a useful checklist for appellate courts. He listed those factors at para. 124. They may be summarized as follows: (a) A previous decision does not reflect the values of the Canadian Charter of Rights and Freedoms; (b) a previous decision is inconsistent with or "attenuated" by a later decision of the court; (c) the social, political, or economic assumptions underlying a previous decision are no longer valid in contemporary society; (d) the previous state of the law was uncertain or a previous decision caused uncertainty; and, (e) in criminal cases, the result of overruling is to establish a rule favourable to the accused.

[114] This list, while useful, is not exhaustive. In the end, this court "will overrule our precedents in an appropriate case" (para. 126). As Laskin J.A. said, at para. 127:

Instead of focusing on phrases such as "manifestly wrong", the approach I prefer is that adopted by this court in R. v. White (1996), 1996 3013 (ON CA), 29 O.R. (3d) 577 (C.A.), at 602. It calls on the court to weigh the advantages and disadvantages of correcting the error in a previous decision. This approach focuses on the nature of the error, and the effect and future impact of either correcting it or maintaining it. In doing so, this approach not only takes into account the effect and impact on the parties and future litigants, but also on the integrity and administration of our justice system.

[115] In my view, this is an appropriate case to overrule the decision in Global Communications, principally because the decision has been overtaken by later decisions of the Supreme Court of Canada dealing with the relationship between freedom of expression and the right to a fair trial. As I have said earlier, Global Communications was premised on the theory that, in a contest between the accused's fair trial rights and freedom of [page124] expression, the accused's fair trial rights were always paramount. I have also noted that this court took the view that the right to a fair trial was a fragile one. The Supreme Court of Canada has rejected the model of Charter analysis based on a hierarchy of rights. As Lamer C.J.C. said in Dagenais, at p. 877 S.C.R.:

The pre-Charter common law rule governing publication bans emphasized the right to a fair trial over the free expression interests of those affected by the ban. In my view, the balance this rule strikes is inconsistent with the principles of the Charter, and in particular, the equal status given by the Charter to ss. 2(b) and 11(d). It would be inappropriate for the courts to continue to apply a common law rule that automatically favoured the rights protected by s. 11(d) over those protected by s. 2(b). A hierarchical approach to rights, which places some over others, must be avoided, both when interpreting the Charter and when developing the common law. When the protected rights of two individuals come into conflict, as can occur in the case of publication bans, Charter principles require a balance to be achieved that fully respects the importance of both sets of rights.

[116] In Dagenais, Lamer C.J.C. also rejected the theory that fair trial rights are so fragile that all possible measures must be taken to protect them. I have set out his comments earlier in these reasons.

[117] Thus, to return to the factors set out in David Polowin, in my view it is necessary to reconsider Global Communications because that decision does not reflect the values of the Charter as they are now understood, and the decision is inconsistent with later decisions of the Supreme Court of Canada, especially Dagenais. I accept that other Polowin factors could favour refusing to overrule Global Communications; for example, the result of overruling arguably establishes a rule less favourable to the accused. However, in my view, the other two factors are so strong in this case that the court should reconsider the earlier decision.

White #2

[118] Following oral argument of this appeal, the Alberta Court of Appeal released its reasons for judgment in White #2. Given the significance of the decision of the Alberta Court of Appeal, this court invited the parties and the intervenors to make written submissions with respect to that decision. We have now received those submissions.

[119] As I have noted above, the media challenged the constitutionality of s. 517(1) in White #2. Brooker J. gave effect to the challenge. He found that the mandatory ban on publication in s. 517 violated s. 2(b) of the Charter and was not a reasonable limit. He held that the words "and shall on application by the [page125] accused" (which create the mandatory ban on publication on application by the accused) were of no force and effect and should be severed from the section. He suspended the declaration of invalidity for 12 months. However, Brooker J. also held that the provision should be amended to open with the words: "Where a jury trial is possible". In other words, neither the Crown nor the accused could apply for a ban on publication under s. 517 where there was no possibility of a jury trial. Brooker J. did not suspend this part of his order.

[120] The Court of Appeal allowed the appeal by the Crown from the judgment of Brooker J., finding that the mandatory limitation on publication provided for in s. 517 was justified under s. 1 of the Charter. The court also considered at length whether Brooker J. should have heard the motion by the media to strike down the legislation. Those same concerns do not arise in this case because of the different context; however, in White #2, as here, it was argued that the Charter issues were largely moot. The Court of Appeal observed, at para. 15, that the case might fall within the category of cases where the court could properly exercise its discretion to hear the motion because the issues are evasive of review.

[121] The court, however, found, at para. 16, that the "most problematic aspect" of Brooker J.'s decision was his declaration that a restriction of publication can never be justified where no jury will be involved. Since White was charged with murder and would be facing a jury trial, the Court of Appeal considered that the issue of a publication ban in non-jury cases was not before Brooker J. The court was of the view that Brooker J. was "faced with a factual vacuum as far as non-jury trials were concerned" (para. 16). The possibility that the mandatory ban under s. 517 could be justified even where a case is tried by a judge alone is referenced several times in the reasons of the Court of Appeal.

[122] Brooker J. found that the Crown had not adduced sufficient evidence to demonstrate a rational connection between the mandatory publication ban provided for in s. 517 and the objectives of the legislation. He took a narrow view of the objective of the legislation, which he found to relate solely to the protection of the accused's right to a fair trial by an impartial jury. He was not satisfied that facilitating the accused's right to an expeditious bail hearing and reducing the burden of litigation on the accused were objectives of the legislation. As I have already discussed, he also had before him expert evidence that pre-trial publicity is unlikely to negatively impact the impartiality of jurors, except in cases where a single piece of information that [page126] essentially decides the cases is given to jurors before a trial but not presented at trial.

[123] The Court of Appeal held that Brooker J. had taken too narrow a view of the objectives of the ban by holding that the law's objective was solely to protect the accused's right to a fair trial by an impartial jury. Slatter J.A., writing for the court, identified, at para. 36, seven additional objectives "directed at preservation of a fair bail hearing and a fair trial", which I would summarize as follows: (a) The mandatory ban is necessary because placing any burden on the accused undermines the presumption of innocence and the right to remain silent. For example, it would be inappropriate to expect an accused to rebut evidence at the bail hearing directed at his character, lifestyle and associates prior to trial. (b) It is unfair to expect the accused to "defend his entitlement to judicial interim release on fair terms against the interests of the media". (c) It is illogical to expect the accused to identify in advance the "specific evidence" that would justify a publication ban, especially since bail hearings are conducted informally without strict application of the rules of evidence and before the Crown has made full disclosure. (d) It would be impossible for the judge to rule on the application without first hearing the evidence; while the judge could make a temporary ban, the accused would have to gamble on whether the ban would be made permanent. (e) Requiring the accused to justify a publication ban could require preparation on the part of the accused and thus delay the bail hearing. (f) The proceedings would be lengthened with the intervention of third parties such as the media and the need to call expert evidence. (g) The expenditure of time and resources for the many bail hearings that are heard each day cannot be justified on a systemic basis.

[124] Having laid out these additional objectives, Slatter J.A. held that the temporary ban on publication until trial was justified. He held that the rational connection test was met when [page127] measured against these other objectives. In doing so, he seems to have concluded that a publication ban could be justified even in non-jury cases. His reasoning is reflected in para. 39, where he concluded that Brooker J. erred in holding that preservation of an untainted jury is the only objective of s. 517:

[Brooker J.'s] analysis also overlooked the other important objectives of s. 517. It precludes consideration of any argument that trial judges (even though they are trained to ignore such matters) should be insulated from pre-trial publicity when possible.

[125] I cannot agree with the Alberta Court of Appeal that an objective of the provision could be to insulate trial judges from pre-trial publicity. Trial judges are expected to ignore such matters; our entire system of non-jury trials is built on the theory that judges can ignore highly prejudicial and inadmissible evidence. Trial judges are routinely asked to rule on the admissibility of confessions, bad character evidence and constitutionally inadmissible evidence. The system depends on the ability of trial judges to ignore evidence deemed inadmissible when reaching the merits of the case. I am not aware of any case that holds that a publication ban can be justified on the basis that a trial judge should not be tainted by pre-trial publicity. Indeed, the Alberta Court of Appeal, when considering a publication ban on an appeal hearing to revoke White's bail, characterized the objective of s. 517 as "aimed at preserving the presumption of innocence and the right to a fair trial by ensuring that prospective jurors have no preconceived notions about guilt prior to trial": R. v. White, 2006 ABCA 65, [2006] A.J. No. 179, 380 A.R. 188 (C.A.), at para. 29 (White #3). While I agree with the Court of Appeal for the reasons set out earlier that the rational connection test has been met, in my view, the rational connection of s. 517 to the objectives cannot be supported by relying on the tainting effect of bail information on non-jury trials.

[126] The Court of Appeal also held that the minimal impairment test has been met. While I have reached the same conclusion, I would comment on one part of the analysis. In considering the minimal impairment test, Slatter J.A. dismissed the media's argument that the newsworthiness of bail information depends upon the information being immediately available. At para. 47, he noted that, since the media decides what is newsworthy, it can always choose to give prominence to the bail proceedings after the trial. He also observed, at para. 49, that the most compelling argument from the media was that s. 517 delays debate on "whether, why and when" accused should be granted bail in high profile cases. He dismissed this argument because "if the media [page128] chose to stimulate a public debate on the place of bail in the criminal justice system, they could easily have done so after the trial, with little overall effect on the long-term public interest because of the delay".

[127] I take a somewhat different view of the nature of the freedom of the press. Even before the enactment of the Charter, the Supreme Court recognized editorial control as an essential attribute of the freedom of the press: Gay Alliance Toward Equality v. Vancouver Sun, 1979 225 (SCC), [1979] 2 S.C.R. 435, [1979] S.C.J. No. 55, at pp. 453-55 S.C.R. As Lamer J. said in Canadian Newspapers Co. v. Canada (Attorney General), supra, at p. 129 S.C.R., measures that prohibit the media from publishing information "deemed of interest" restrict freedom of the press. Freedom of the press encompasses not just the freedom to decide what to report but also the timing of the reporting. If, by the time the trial is finished, what transpired at the bail hearing is no longer deemed newsworthy, the Canadian public as consumers of the news are the poorer for it. Immediacy is the essence of news: Dagenais, at p. 929 S.C.R., per Gonthier J. (dissenting). To repeat what Doherty J.A. said in Domm, at para. 40, the values promoted by s. 2(b) are not served when "the message to be conveyed has lost its purpose". It is not for Parliament or the courts to dictate the timing of debate on issues of public importance, but that is exactly the effect of the temporary ban in s. 517.

[128] As to the final stage of the s. 1 analysis, whether the salutary effects of the legislation are proportionate to its deleterious effects, Slatter J.A. considered that s. 517 struck the necessary balance between the rights to a fair trial and reasonable bail and freedom of the press. As he said, at para. 57 [R. v. White (C.A.)]"the benefits of the restrictions on publication in s. 517 outweigh the negative effects of those restrictions". I now wish to return to a discussion of the objectives identified by the Alberta Court of Appeal. It seems to me that the principal reason that court reached a different conclusion than I have on this final factor is because of the broad view the court took of the objectives of the mandatory ban.

[129] As is apparent from my reasons, I agree with several of the objectives identified by the Alberta Court of Appeal, particularly those identified as objectives (c) and (d). However, I cannot agree with several of the other objectives. For the reasons expressed earlier, I cannot agree that preventing the tainting of a trial judge in a judge-alone trial is a valid objective.

[130] As well, I cannot agree that placing any burden on the accused to justify a limitation on publication undermines the [page129] presumption of innocence and the right to remain silent (objective (a)). The fact that, for tactical reasons, the accused may feel it necessary to adduce evidence at the bail hearing does not undermine either the presumption of innocence or the right to remain silent. The Supreme Court of Canada has spoken on this issue, albeit in a different context, in R. v. Darrach, 2000 SCC 46, [2000] 2 S.C.R. 443, [2000] S.C.J. No. 46, 148 C.C.C. (3d) 97. In that case, the Supreme Court upheld the procedure the accused must follow to lay the groundwork for an application to adduce evidence of the complainant's prior sexual conduct under s. 276 of the Criminal Code. In that case, the court held that requiring an accused to adduce evidence, which may include disclosure of his defence to a sexual assault charge, does not infringe the Charter. Given the decision in Darrach, I fail to see how the accused's rights are infringed by the need to demonstrate why there should be a publication ban. An accused seeking a publication ban under s. 517 is not forced to testify, nor is the accused coerced by the state.

[131] The Alberta Court of Appeal held that it was inappropriate to expect an accused to rebut evidence of character, lifestyle and associates prior to his trial. That may be so, but I fail to see why this justifies a mandatory publication ban. In Hollinger Inc. v. Ravelston Corp. (2008), 2008 ONCA 207, 89 O.R. (3d) 721, [2008] O.J. No. 1126 (C.A.), leave to appeal to S.C.C. refused [2008] S.C.C.A. No. 260, Juriansz J.A., dissenting in part, noted, at para. 97, that the open court principle would be slender if it did not apply to documents that had not been tested as to their factual probity. Similarly, the open court principle and the right to freedom of expression would be weak if the media could only report on information after it had been fully tested and contested at a full trial. As well, the mandatory ban on bail hearing proceedings does not prevent the publication of untested and unfiltered material that comes into the hands of the media from other sources, such as the pre-bail-hearing police press- conference.

[132] The Alberta Court of Appeal also considered that it was unfair to expect the accused, who is frequently unrepresented at the bail hearing, to defend his entitlement to bail on fair terms against the interests of the media (objective (b)). In my view, seeking a discretionary ban does not implicate the accused's entitlement to bail. The media is not a party to the bail hearing and has no say in the accused's entitlement to bail. The accused's only opponent on the issue of bail is the prosecution.

[133] As to objectives (e) and (f), concerning the problem of delay of the hearing, I have already indicated that, in my view, the hearing need not be delayed to give notice to the media. [page130]

[134] Finally, the Alberta Court of Appeal was of the view that the expenditure of time and resources that would be required when adjudicating applications for discretionary bans cannot be justified on a systemic basis (objective (g)). The court, however, did not refer to any evidence that would support this proposition. It would seem to me that, since the court is required to consider the actual deleterious and salutary effects, speculation about resources is not sufficient. Remedy

[135] Section 52 of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11 provides that any law that is inconsistent with the Constitution is, to the extent of the inconsistency"of no force and effect". In the circumstances of this case, since it is possible to define with precision the inconsistent part of the provision, the appropriate remedy would seem to be severance by removing the words "and shall on application by the accused" that create the mandatory publication ban in s. 517. This would leave in place the discretionary ban. In deciding whether to exercise that discretion, the justice or judge would be guided by the Dagenais directives: see Canadian Broadcasting Corp., at paras. 68-76.

[136] The only other remedy would be to strike down the whole of s. 517. In choosing between these two remedies, the principle considerations appear to be those set out by Sopinka J. in Osborne v. Canada (Treasury Board), 1991 60 (SCC), [1991] 2 S.C.R. 69, [1991] S.C.J. No. 45, at p. 104 S.C.R., being the need to fashion a remedy to "apply the measures which will best vindicate the values expressed in the Charter" and "refrain from intruding into the legislative sphere beyond what is necessary": see Schachter v. Canada, 1992 74 (SCC), [1992] 2 S.C.R. 679, [1992] S.C.J. No. 68, at p. 707 S.C.R.

[137] In my view, severance or reading down is the appropriate remedy. This is not a case like R. v. Seaboyer, 1991 76 (SCC), [1991] 2 S.C.R. 577, [1991] S.C.J. No 62, 66 C.C.C. (3d) 321, where the court rejected reading down the impugned provision to provide for discretion because Parliament had specifically chosen to exclude that element. Section 517 has always contained a discretionary element and the mandatory ban would in effect restore the provision to its original form when the Bail Reform Act was enacted. I do not therefore see severance as a substantial change to the law.

[138] As Lamer C.J.C. pointed out in Schachter, at p. 711 S.C.R., another way of testing the appropriateness of the remedy is to look at the effect on the remaining portion. As he said"The [page131] problem with striking down only the inconsistent portion is that the significance of the remaining portion changes so markedly without the inconsistent portion that the assumption that the legislature would have enacted it is unsafe." In my view, that cannot be said in this case.

[139] I am strengthened in this view when the issue is looked at from the point of view of the "significance or long-standing nature of the remaining portion": Schachter, at p. 712 S.C.R. As Lamer C.J.C. said, at p. 713 S.C.R. of Schachter: "If the remaining portion is very significant, or of a long standing nature, it strengthens the assumption that it would have been enacted without the impermissible portion." Given the fact that the provision as originally enacted was solely a discretionary ban, I am satisfied that severance is preferable to striking down the entire provision and is the remedy that least interferes with the intention of Parliament.

[140] Accordingly, I would allow the appeal from the application judge's 2007 decision and hold that the words "and shall on application by the accused" in s. 517 of the Criminal Code are of no force and effect.

[141] The next issue to consider is whether the declaration of invalidity should be suspended. In Schachter, at p. 719 S.C.R., Lamer C.J.C. explained the circumstances in which it is appropriate to temporarily suspend the declaration of invalidity:

Temporarily suspending the declaration of invalidity to give Parliament or the provincial legislature in question an opportunity to bring the impugned legislation or legislative provision into line with its constitutional obligations will be warranted even where striking down has been deemed the most appropriate option on the basis of one of the above criteria if:

A. striking down the legislation without enacting something in its place would pose a danger to the public;

B. striking down the legislation without enacting something in its place would threaten the rule of law; or,

C. the legislation was deemed unconstitutional because of underinclusiveness rather than overbreadth, and therefore striking down the legislation would result in the deprivation of benefits from deserving persons without thereby benefitting the individual whose rights have been violated.

[142] Lamer C.J.C. went on to point out that these propositions were intended only as guidelines. As Brooker J. noted in White #2, at para. 124, since Schachter, the Supreme Court has suspended the declaration of invalidity in freedom of expression cases not falling within these guidelines where the legislation was overbroad. [page132]

[143] In White, Brooker J. found that criterion B was met. He wrote, at para. 118, as follows:

In my view, to remove the mandatory portion of s. 517 would offend the Rule of Law and result in a legislative gap. This is because, in light of the current notice requirements to the media and restrictions on bail hearing adjournments, such a severance without replacement legislation would have the effect of potentially denying accused persons their constitutional right not to be arbitrarily detained under s. 9 of the Charter, which forms part of the fundamental law of this country.

[144] For somewhat different reasons, I too would suspend the declaration of invalidity. I am not convinced that the impact of removing the mandatory ban would be profound given that the deleterious effects outweigh the salutary effects. Further, as I have indicated, it is my view that it is not necessary for the bail hearing to be delayed to give notice to the media. Thus, I do not see the danger of infringement of the accused's rights to protection against arbitrary detention.

[145] However, I am concerned that removing the mandatory ban and leaving in place only the discretionary ban in s. 517 may not be sufficient to meet other objectives that are currently served by the mandatory ban. I return to the wording of s. 517, having removed the words which allow for the mandatory ban:

517(1) If the prosecutor or the accused intends to show cause under section 515, he or she shall so state to the justice and the justice may, before or at any time during the course of the proceedings under that section, make an order directing that the evidence taken, the information given or the representations made and the reasons, if any, given or to be given by the justice shall not be published in any document, or broadcast or transmitted in any way before such time as (a) if a preliminary inquiry is held, the accused in respect of whom the proceedings are held is discharged; or (b) if the accused in respect of whom the proceedings are held is tried or ordered to stand trial, the trial is ended.

[146] My principal concern is that, as worded, the discretionary ban that can be imposed under s. 517 applies to all information relating to the proceedings. Put another way, a justice of the peace might not have the jurisdiction under s. 517 to impose a carefully tailored ban. For instance, a limited prohibition on the publication of the accused's confession might be the only type of ban which passes constitutional muster, yet the justice of the peace might not be able to impose such a ban.

[147] It may be that a justice of the peace has jurisdiction at common law to make a carefully tailored order, but the jurisdiction of a statutory court to resort to the common law in the face of [page133] an express statutory provision is unclear: see the full discussion of the jurisdiction of courts of record to impose publication bans in Church of Scientology of Toronto and the Queen (No. 6) (Re), 1986 4649 (ON SC), [1986] O.J. No. 3068, 27 C.C.C. (3d) 193 (H.C.J.). In other words, I am concerned that striking out the mandatory ban may leave a gap because a justice of the peace would not have the jurisdiction either under the revised s. 517 or at common law to make the kind of order that is necessary and would meet constitutional standards.

[148] For a discretionary ban to be justified, it must meet the Dagenais/Mentuck test. I have set out those tests earlier but repeat the test from para. 32 of Mentuck here for convenience:

A publication ban should only be ordered when:

(a) such an order is necessary in order to prevent a serious risk to the proper administration of justice because reasonably alternative measures will not prevent the risk; and

(b) the salutary effects of the publication ban outweigh the deleterious effects on the rights and interests of the parties and the public, including the effects on the right to free expression, the right of the accused to a fair and public trial, and the efficacy of the administration of justice.

[149] The test for making a publication ban where authorized by statute was framed in similar terms in Canada Broadcasting Corp. v. New Brunswick (Attorney General). The court held, at para. 69, that it is necessary to:

(a) . . . consider the available options and consider whether there are any other reasonable and effective alternatives available;

(b) . . . consider whether the order is limited as much as possible; and

(c) . . . weigh the importance of the objective of the particular order and its probable effects against the importance of openness and the particular expression that will be limited in order to ensure that the positive and negative effects of the order are proportionate.

[150] A justice of the peace might be persuaded that there was justification for a publication ban at common law or under statute in accordance with these tests, but only in respect of certain information. For example, an accused might be able to show through submissions that he or she will be denied the right to reasonable bail because sureties will refuse to come forward if the proceedings are published. The justice of the peace might be satisfied that the Dagenais/Mentuck test can only be met through a narrowly tailored ban, for example, a ban on publication of the identity of the sureties. Similarly, where the prosecution intends to adduce the fact that the accused confessed, the justice of the peace might be convinced the Dagenais/Mentuck [page134] test had been met, but only in respect of publication of the confession and not the entire proceedings, including the reasons for judgment. Parliament may wish to consider whether it is necessary to amend s. 517 so that a justice of the peace or judge can impose a limited publication ban.

[151] At the same time, it is unclear whether a justice of the peace acting under Part XVI of the Criminal Code has jurisdiction to make a common-law publication ban. While the courts have recognized that trial judges and superior courts have jurisdiction to impose publication bans, it is not apparent that a justice of the peace or judge of the Ontario Court of Justice conducting a bail hearing has that jurisdiction. In Dagenais, at p. 857 S.C.R., Lamer C.J.C. wrote as follows:

To seek a ban under a judge's common law or legislated discretionary authority, the Crown and/or the accused should ask for a ban pursuant to that authority. This request should be made to the trial judge (if one has been appointed) or to a judge in the court at the level the case will be heard (if the level of court can be established definitively by reference to statutory provisions such as ss. 468, 469, 553, 555, 798 of the Criminal Code, R.S.C., 1985, c. C-46, and s. 5 of the Young Offenders Act, R.S.C., 1985, c. Y-1). If the level of court has not been established and cannot be established definitively by reference to statutory provisions, then the request should be made to a superior court judge (i.e., it should be made to the highest court that could hear the case, in order to avoid later having a superior court judge bound by an order made by a provincial court judge). (Emphasis added)

[152] It would be highly impractical to require an accused to apply to the Superior Court of Justice for a publication ban; the jurisdiction to make such an order should reside in the court conducting the show cause hearing. I should not be taken as having decided the issue of whether a justice of the peace or judge conducting a show cause hearing has the necessary common-law jurisdiction. However, given the uncertainty surrounding the issue and the possibility of a gap in the legislation, the declaration of invalidity should be suspended to give Parliament the opportunity to clarify the jurisdiction to make the order.

[153] Parliament may also wish to consider enacting a provision to complement s. 542 of the Criminal Code, which prohibits publication of a report that any admission or confession was tendered in evidence at a preliminary inquiry or a report of the nature of such admission or confession.

[154] Parliament may also wish to enact legislation to clarify the bases for making a publication ban. In Mentuck, the Supreme Court held that a publication ban can be justified for reasons other than preserving the accused's right to a fair trial. A publication ban could be imposed in the interests of the [page135] administration of justice. I would think it open to Parliament to provide guidance to bail courts by setting out in legislation the circumstances that would justify a discretionary publication ban in the interests of the administration of justice.

[155] Finally, Parliament may wish to enact measures to clarify the procedure for obtaining a discretionary publication ban.

[156] For these reasons, I would suspend the declaration of invalidity to give Parliament an opportunity to amend the legislation, if so inclined. Disposition

[157] Accordingly, I would allow the appeal from the order of Durno J. dated March 1, 2007 and declare the words "and shall on application by the accused" in s. 517 of the Criminal Code to be of no force and effect. I would suspend the declaration of invalidity for 12 months. I would dismiss the appeal from the order of Durno J. dated July 27, 2006. Some of the respondents and interested parties sought costs. This is not a case for costs.

FELDMAN J.A. (LASKIN and SIMMONS JJ.A. concurring): -- Introduction

[158] I have had the benefit of reading the full and careful reasons prepared by my colleague, Rosenberg J.A., where he concludes that the mandatory publication ban of bail proceedings when requested by an accused, contained in s. 517 of the Criminal Code, R.S.C. 1985, c. C-46, violates s. 2(b) of the Canadian Charter of Rights and Freedoms, freedom of the press, and is not saved by s. 1. He finds that the provision meets the pressing and substantial objective, the rational connection and the minimal impairment criteria under Oakes [1986 46 (SCC), [1986] 1 S.C.R. 103, [1986] S.C.J. No. 7], and bases his conclusion only on the third branch of the proportionality test, that in seeking to protect the accused's fair trial rights, the salutary effects of the mandatory ban do not outweigh its deleterious effects. He would strike down the mandatory publication ban, but suspend the order for 12 months to allow Parliament to devise and provide a scheme whereby judges and justices of the peace would have the discretion to impose limited bans in particular cases.

[159] I agree that the mandatory publication ban, as currently drafted, violates s. 2(b) of the Charter, and is not saved by s. 1. However, in my view, this is because it is overly broad in its scope, applying to bail hearings for all charges, including those that cannot be tried by a jury; the provision, therefore, to that [page136] extent, does not meet the rational connection or the minimal impairment tests under Oakes.

[160] A publication ban is needed in jury cases to protect an accused's right to a fair trial under s. 11(d) and s. 7 of the Charter by preventing potential jurors from learning of prejudicial information from bail hearings which may never be heard at trial. In my view, because it is also critically important that the rights of an accused under s. 11(e) and s. 7 of the Charter to reasonable bail following an expeditious bail hearing be protected in possible jury cases, the publication ban must be mandatory at the request of an accused and not merely available at the discretion of the justice following a hearing on that issue. I would therefore leave the mandatory ban in place, but read down the legislation to make it applicable only to show cause hearings in respect of charges that procedurally may possibly be tried by a jury. Preliminary Issues

[161] Rosenberg J.A. has dealt with a number of preliminary issues raised on this appeal, including (1) that this appeal should be heard regardless of whether it is moot; (2) that the court will consider the fresh evidence tendered by the Attorney General of Ontario; (3) that the trial judge was correct in his interpretation of s. 517 that the mandatory publication ban applies to all jointly charged accused when one accused asks for it; and (4) that it is appropriate for this court to reconsider its 1984 decision in Global Communications Ltd. and Canada (Attorney General) (Re) (1984), 1984 2153 (ON CA), 44 O.R. (2d) 609, [1984] O.J. No. 3066 (C.A.), where the court upheld the constitutional validity of the mandatory publication ban under s. 517. I agree with his analysis and his disposition of these issues. The Charter Issues

[162] It is accepted that the s. 517 mandatory publication ban on bail hearings at the request of an accused violates s. 2(b) of the Charter. The issue before this court is whether and to what extent it is saved under s. 1. I am in full agreement with much of my colleague's analysis. I agree with his conclusion that the legislation has a pressing and substantial objective. In my view, as currently worded, the legislation does not pass the rational connection and minimal impairment stages of the Oakes test insofar as it mandates a bail hearing publication ban where a future jury trial is not possible. In all other respects, I agree that the provision meets both the rational connection and the [page137] minimal impairment requirements of the Oakes test. Finally, once the mandatory ban is limited to situations where a future jury trial is possible, I do not agree with Rosenberg J.A. that the legislation fails the third part of the proportionality test comparing salutary and deleterious effects. In my view, the salutary effects of the mandatory publication ban outweigh its deleterious effects, and the provision is saved under s. 1 of the Charter. Pressing and Substantial Objective

[163] As Rosenberg J.A. states, the parties appear to disagree on the objective of the legislation. However, that apparent disagreement stems from the fact that the parties are not separately addressing the two components of the legislation: (1) the publication ban itself, and (2) the fact that it is mandatory when requested by the accused. Each of these components has its own objective. In my view, in order to conduct an effective s. 1 analysis, the two components of the legislation should be addressed separately at each stage. For ease of reference, s. 517 reads:

517(1) If the prosecutor or the accused intends to show cause under section 515, he or she shall so state to the justice and the justice may, and shall on application by the accused, before or at any time during the course of the proceedings under that section, make an order directing that the evidence taken, the information given or the representations made and the reasons, if any, given or to be given by the justice shall not be published in any document, or broadcast or transmitted in any way before such time as (a) if a preliminary inquiry is held, the accused in respect of whom the proceedings are held is discharged; or (b) if the accused in respect of whom the proceedings are held is tried or ordered to stand trial, the trial is ended.

(i) The purpose of the publication ban itself

[164] The purpose of imposing a publication ban on bail proceedings is to protect the accused's right to a fair trial by an impartial jury, a right guaranteed by ss. 7 and 11(d) of the Charter. The accused's right to an impartial jury, and therefore to a fair trial, may be undermined if members of the jury have heard or read about details from the bail proceedings that may not be heard at the trial. Section 518 of the Code gives broad latitude for the introduction of evidence at the bail hearing that either may not be admissible at the trial or may be presumptively inadmissible at the trial, including wiretaps, the accused's complete criminal record, other pending charges, [page138] confessions and statements by the accused and hearsay that the justice considers credible or trustworthy.

[165] The potential threat to trial fairness was recognized by this court in Global Communications and by the Supreme Court of Canada in Dagenais v. Canadian Broadcasting Corp., 1994 39 (SCC), [1994] 3 S.C.R. 835, [1994] S.C.J. No. 104, at pp. 884-86 S.C.R. The extent of the vulnerability of juries to tainting is clearly a matter of ongoing debate both in the courts and in social science empirical research and analysis. For example, up until this court's decision in R. v. Parks (1993), 1993 3383 (ON CA), 15 O.R. (3d) 324, [1993] O.J. No. 2157, 84 C.C.C. (3d) 353 (C.A.), leave to appeal to S.C.C. refused (1993), 17 O.R. (3d) xvii, [1993] 1 S.C.R. x, [1993] S.C.C.A. No. 481, 87 C.C.C. (3d) vi, although there was some social science evidence to the contrary, it was accepted by our courts that white jurors could be trusted to put aside any racial prejudice they may have harboured against a black accused in order to carry out their duties as impartial arbiters in accordance with the instructions of the trial judge. Unfortunately, once the challenge for cause was extended to the issue of racial prejudice following Parks, it became apparent that a number of jurors were not able to set aside their prejudice and decide the case only according to the evidence and the instructions of the trial judge.

[166] The discussion of whether and to what extent juries may be affected by or act on information they may learn outside the trial and therefore the effectiveness of a publication ban on bail proceedings could be undertaken under the rational connection test. However, because it is the main area that affected Rosenberg J.A.'s conclusion that the salutary effects of the ban are weak and speculative, I prefer to address this important point as part of the analysis of the pressing and substantial nature of the objective of the provision.

[167] In his text, The Law of Bail in Canada, 2nd ed. (Toronto: Carswell, 1999), Justice Trotter identifies some recent social science writing which suggests that juries are in fact more influenced by information they glean outside the trial than was previously believed. At this point, we have no definitive evidence of the extent to which juries' verdicts may be affected by such information. What we do have are common-law rules developed by the courts to address the issue and to try to ensure, to the extent possible, that juries will not allow any such information to affect their verdicts.

[168] The continued existence of these rules reflects the ongoing concern of the courts and the common law that jurors may, whether advertently or inadvertently, allow their judgment to be [page139] influenced by prejudicial information that was not part of the evidence at the trial. [See Note 9 below] Examples of prejudicial evidence that is kept from juries include involuntary confessions or confessions made in breach of an accused's Charter rights, criminal convictions closely related to the charge that may be excluded on an R. v. Corbett, 1988 80 (SCC), [1988] 1 S.C.R. 670, [1988] S.C.J. No. 40, 41 C.C.C. (3d) 385 application, similar fact evidence that does not meet the test in R. v. Handy (2002), 2002 SCC 56, 61 O.R. (3d) 414, [2002] 2 S.C.R. 908, [2002] S.C.J. No. 57, 164 C.C.C. (3d) 481, bad character evidence if the accused does not put his or her character in issue, and hearsay evidence that is excluded because it is not sufficiently necessary or reliable to be admissible.

[169] Other evidence is admitted in most cases with a specific instruction to the jury as to its use and misuse and with an explanation for the limitations. Examples include the evidence of an accomplice or jail house informant where a Vetrovec warning (R. v. Vetrovec, 1982 20 (SCC), [1982] 1 S.C.R. 811, [1982] S.C.J. No. 40, 67 C.C.C. (2d) 1) explains to the jury why such a witness may have a reason to lie, similar fact evidence that meets the Handy test and is allowed to be heard and properly considered by the jury in relation to specific issues, prior inconsistent statements that are not adopted by the witness, and after-the-fact conduct. As well, the jury is always told by the presiding judge that the questions and submissions of counsel that they hear in court are not evidence and cannot be taken as evidence unless they are adopted by a witness or a document that is in evidence.

[170] Juries are also directed by the trial judge to base their decision only on the evidence presented in court and to put out of their minds anything they have heard, seen or read about the case outside the courtroom. However, that general instruction is not the equivalent of completely excluding prejudicial evidence, or of coupling its admission with specific directions that identify the evidence and describe and explain its permitted uses.

[171] While maintaining the common-law and Charter rules for the exclusion of certain evidence and the importance of limiting instructions by trial judges, the Supreme Court of Canada [page140] has repeatedly emphasized its trust in juries to follow the instructions of the trial judge. The strength of our jury system depends on the ability and the willingness of jurors not only to apply their common sense when considering the evidence, but also to take the law from the judge and to act in accordance with the judge's directions regarding their duties and obligations. In cases such as Corbett and Dagenais, the Supreme Court has struggled with how to give full effect to that trust while at the same time ensuring that an accused person will have a fair trial by an impartial jury.

[172] In Corbett, for example, Dickson C.J.C., for the majority, considered s. 12 of the Canada Evidence Act, R.S.C. 1970, c. E-10, which allows a witness, including an accused, to be cross-examined on his or her criminal record. He concluded that the provision did not violate the presumption of innocence in s. 11(d) of the Charter because the jury is instructed that it can only use the record for the issue of credibility and not as proof of guilt. He noted that other safeguards, such as the prohibition on cross-examination of the accused on the underlying circumstances of the convictions, assist the jury in understanding that the only reason they are hearing about the record is to give them a full picture of the accused person's credibility. Finally, Dickson C.J.C. found that in order to ensure the accused's fair trial rights, the trial judge retains the discretion under s. 12 to, in effect"edit" the accused's criminal record where any probative value regarding the accused's credibility would be outweighed by the prejudicial effect of a prior conviction, particularly for a similar offence. He stated, at p. 697 S.C.R., p. 404 C.C.C.:

These limitations on the use of prior convictions, together with the discretion recognized by the reasons of La Forest J., demonstrate a marked solicitude for the right of the accused to a fair trial and indicate that the law relating to the use of prior convictions strives to avoid the risk of prejudicing an accused's trial by introduction of evidence of prior misdeeds. Taken as a whole, this body of law is entirely protective of the right of the accused not to be convicted except on evidence directly relevant to the charge in question. Within this context, it cannot be said that s. 12 of the Canada Evidence Act operates in such a way as to deprive the accused of the right to a fair trial.

[173] In Dagenais, Lamer C.J.C. also expressed his trust in the ability of jurors to follow the instructions of the trial judge even in cases where they will have been exposed to pre- trial publicity. However, the thrust of his reasons still focused on ways other than a publication ban for the court to address the fair trial concern created by the undeniable potential for [page141] the jury to be affected by pre-trial publicity. He stated, at p. 887 S.C.R.:

[A]lthough it is possible that a publication ban will have a total absence of influence on the fairness of the trial, such cases will be rare. As a result, one will generally have to go further and consider the availability of reasonable alternative measures when assessing whether, in a given case, a publication ban was necessary. (Emphasis in original)

[174] Rosenberg J.A. concluded that the fair trial objective of the s. 517 publication ban is pressing and substantial and that the measure is rationally connected to its purpose, to protect the fair trial rights of an accused against possible jury contamination. I agree with his conclusion. However, later in his proportionality analysis, his conclusion that the salutary effects of the mandatory ban are outweighed by its deleterious effects is based in large part on his concern that the causal connection between pre-trial publicity and jury contamination is weak and speculative and his acceptance that jury members would not remember most if not all of such publicity by the time of the trial. I do not share that concern nor, do I accept that view.

[175] For the reasons I have outlined above, I conclude that the potential for jurors to be affected by prejudicial information they have heard outside the trial has been accepted by the courts as something that must be guarded against in order to protect both the accused's right to a fair trial and the public's right to be confident that justice is done. Clearly some jurors will be able to follow their oaths and disabuse themselves of prior information they may have learned about the accused or the case. Our faith in juries is based on their willingness to do so. However, some jurors may not be able to do so, or may not realize that they have been influenced or affected by pre-trial publicity. Further, in some instances, depending on the nature of the prejudicial information, it may be impossible for even the most conscientious jurors to disabuse themselves of that information. Sometimes, in order to prevent prejudice to the accused, the proper remedy is the declaration of a mistrial: see R. v. Burke, 2002 SCC 55, [2002] 2 S.C.R. 857, [2002] S.C.J. No. 56, 164 C.C.C. (3d) 385, at para. 74, where Major J. stated that one circumstance where a mistrial may be declared to prevent a miscarriage of justice is "where inadmissible evidence is adduced during trial which might influence a jury".

[176] The common law has developed rules as well as discretionary protocols for trial judges to use to ensure that jurors either do not hear certain evidence, or if they do hear it, they are [page142] clearly instructed what use they can make of it. It is the application of these rules and protocols, together with other trial orders such as change of venue, that allow us to be satisfied that the accused is receiving a fair trial by an impartial jury.

[177] It is also, in my view, no longer appropriate or realistic to rely on jurors' faded memories of any pre-trial publicity by the time of the trial as the basis for confidence that they will not remember what they read or heard. Once something has been published, any juror need only "Google" the accused on the Internet to retrieve and review the entire story.

[178] I am satisfied that the purpose of the s. 517 publication ban on bail hearings is pressing and substantial based on the acknowledged potential for jury contamination. (ii) The purpose of the publication ban being mandatory at the request of the accused

[179] The purpose of the publication ban on bail proceedings being mandatory at the request of the accused is to protect the accused's s. 11(e) Charter right to reasonable bail, as well as the accused's s. 7 liberty interest in expeditious bail proceedings. Because the order is mandatory, there is no need for adjournments that may leave the accused waiting in custody, to give notice to the media or to give the accused the opportunity to retain appropriate counsel to prepare and conduct an application for a discretionary ban.

[180] Similarly, because the ban is mandatory, accused know that a ban will be ordered. Therefore, they do not need to be concerned that information presented at the bail hearing may be exposed in the press and they need not plan their bail cases with that risk in mind. Nor need an accused, either personally or through the legal aid system, expend financial resources to obtain a publication ban, resources which might otherwise be spent on obtaining bail and preparing for trial.

[181] Rosenberg J.A. characterized these issues as forming part of an expanded view of preserving an accused's and society's interest in a fair trial. I agree with his statement, in para. 38 of his reasons, that:

The objectives of ensuring expeditious bail hearings, avoiding unnecessary detention of accused and allowing accused to retain scarce resources to defend their cases are all inextricably linked to the objective of ensuring a fair trial.

[182] I am also in agreement that this objective is pressing and substantial. [page143] The Proportionality Test

Rational connection (i) The publication ban itself

[183] The rational connection test asks the court to consider whether the infringement, in this case, the infringement on freedom of the press by a mandatory publication ban, is rationally and logically connected to the legislation's objective: see RJR-MacDonald Inc. v. Canada (Attorney General), 1995 64 (SCC), [1995] 3 S.C.R. 199, [1995] S.C.J. No. 68, at para. 153. As I believe it is necessary to break down the legislative objective into its two components (the purpose of the ban and the purpose of making it mandatory), I am considering under this subheading the objective of the ban itself, preserving the fair trial rights of the accused.

[184] In my view, for the reasons discussed under the "pressing and substantial objective" heading, a publication ban on bail proceedings is rationally connected to the preservation of an accused's right to a fair trial by jury, but not by a judge sitting alone.

[185] Fair trial rights cannot be said to be at risk where a judge, sitting alone, is exposed to prejudicial information which should not be admitted at trial. Judges are effectively professional decision-makers. They know the rules they must apply when considering the admissibility of different types of evidence. They are required and are able to disabuse themselves of prejudicial information they may learn in a voir dire setting or in the context of other mid- or pre-trial rulings they are routinely called upon to make. If they were not trusted to do so, then our trial system would not be able to function as it does.

[186] Later in these reasons, I will discuss the remedy for the overbreadth of the provision. (ii) The mandatory nature of the bail publication ban

[187] I will address the balance of the issues under the proportionality test on the assumption that the mandatory ban applies only to bail hearings where the accused is charged with an offence that is potentially triable by jury.

[188] Because the ban is mandatory at the request of the accused, s. 517 creates "an expeditious and effective means" of obtaining the order (see Rosenberg J.A.'s reasons at para. 50). Because the ban is mandatory and not discretionary, it guarantees that (1) the accused's fair trial interest will not be violated; (2) the accused will not suffer delay in obtaining bail, i.e., his or her liberty; [and] (3) the accused will not have his or her financial [page144] and other resources diverted from the bail and trial preparation processes to fight against the media for a publication ban.

[189] If the publication ban issue were discretionary, whether the media were present or not, in every bail proceeding where a publication ban was requested the justice would be required to consider the propriety of making the order. The appellants argued, and Rosenberg J.A. accepted, that a justice would have to conduct a Dagenais analysis in each case in order to determine whether to order a ban in the particular case.

[190] There are several problems with attempting to apply the Dagenais test in the bail hearing context when an accused is seeking to protect his or her fair trial rights. To begin, Dagenais involved a request for a publication ban based on common-law principles in order to prevent the airing of a television show that dramatized very unsavoury events relating to cases that were being heard or were about to be heard before juries. Lamer C.J.C. emphasized that the principles discussed applied to a common-law, not a statutory, publication ban order. The context was an extraordinary set of circumstances that arose close to the trial, not an everyday procedure that occurs at the very beginning of the criminal process before the accused has been given disclosure or the time to evaluate the case with counsel and to make important and informed decisions about procedural and substantive issues.

[191] Lamer C.J.C. held that the application must go before a superior court judge who has the discretion to tailor an order to meet the circumstances of the case. Importantly, Lamer C.J.C. held that the judge should turn first to alternative methods of preventing the jury from being affected by adverse pre-trial publicity, alternative methods such as a change of venue, adjournment of the trial, sequestering jurors, challenge for cause or strong judicial direction to the jury.

[192] It is apparent that the Dagenais process does not transfer easily and effectively to the ordinary bail proceeding and that Lamer C.J.C. did not intend the process he described to be applied by a justice of the peace on a regular basis in an everyday hearing that is intended to be expeditious. [See Note 10 below] As Lamer C.J.C. noted, at p. 891 S.C.R., the Dagenais hearing involves a complex constitutional weighing where the accused who is seeking bail would have: [page145]

. . . the burden of proving that the proposed ban is necessary, in that it relates to an important objective that cannot be achieved by a reasonably available and effective alternative measure, that the proposed ban is as limited (in scope, time, content, etc.) as possible, and there is a proportionality between the salutary and deleterious effects of the ban. At the same time, the fact that the party seeking the ban may be attempting to safeguard a constitutional right must be borne in mind when determining whether the proportionality test has been satisfied.

[193] Even if an accused were able to meet this heavy legal burden, it is clearly unrealistic to expect that accused persons in the position I have described could do so within any form of expeditious time frame.

[194] Further, the presiding judicial officer is required to first consider all other options, then all ways to limit the ban, if it is going to be ordered, and must weigh "the importance of the objectives of the particular ban and its probable effects against the importance of the particular expression that will be limited to ensure that the positive and negative effects of the ban are proportionate" (p. 891 S.C.R.).

[195] If it is accepted, as Rosenberg J.A. did, that the mandatory ban does not pass constitutional muster because the concern for fair trial rights based on jury tainting is generally a speculative concern, I find it difficult to imagine in what circumstances a discretionary ban on bail proceedings would ever be ordered to protect an accused's fair trial rights.

[196] Nor can a justice of the peace make any order that affects the trial, such as an adjournment or change of venue order. These types of alternatives are available to a superior court judge who deals with the issue of prejudice just before trial, but not to a justice on a bail hearing. Rosenberg J.A. has also raised doubts whether a justice of the peace currently has the authority to tailor or limit a ban. The result is that, while the Dagenais analysis would require the justice to consider alternative measures to protect the fair trial rights of the accused, the justice would have no control or formal input into whether any such alternatives are actually implemented. This also creates further uncertainty for the accused when making the decision whether to seek bail.

[197] Along with the issues regarding the suitability of the Dagenais procedure and test to the bail hearing process, the amount of time needed to conduct a discretionary publication ban hearing in every case where an accused asks for a discretionary ban would further overburden an already overburdened bail system. Whether opposed or unopposed, the motion for the ban would add to the length of the bail hearing. At least initially, criminal duty counsel may feel unprepared to properly conduct a [page146] publication ban request proceeding according to the Dagenais criteria and likely will require extra time to prepare. There will be a need for adjournments and hearings will be longer.

[198] In Ontario, the current problem with overburdened bail courts was recently highlighted by the decision of De Filippis J. in R. v. Jevons, [2008] O.J. No. 4397, 2008 ONCJ 559, where he stayed criminal charges on the date fixed for trial because of the delay in the accused's bail proceeding. In his reasons, De Filippis J. gave a detailed account of the accused's eight- day ordeal waiting for a court to have the time to hear his bail application. In his affidavit on the stay motion, the accused testified that when the Crown finally agreed to consent to bail, he was "prepared to agree to just about anything to secure...release" and he had "serious doubts" that he would ever obtain a bail hearing (para. 13).

[199] De Filippis J. also described the backlog plaguing the bail courts in the Durham Region and the fact that, for the week in question, the court was unable to deal with the majority of contested bail applications and only had time for consent bails and adjournments. De Filippis J. concluded that the accused was the victim of systemic delay in the bail system in Durham Region and that his Charter right to reasonable bail under s. 11(e) was infringed, as was his liberty interest under s. 7.

[200] In summary, in order to be of practical value to an accused seeking to protect his or her fair trial right through a publication ban on the bail proceedings, the ban must be mandatory and not discretionary. The existing procedure for seeking a discretionary ban under Dagenais puts a heavy and difficult onus on the accused and, if implemented following Rosenberg J.A.'s analysis, appears to have little likelihood of success. In fact, counsel for the appellants acknowledged that very few discretionary bans would likely be granted and eventually accused persons seeking bail would not ask for publication bans, thereby solving any delay and cost problems associated with seeking the discretionary order.

[201] The consequence could be that accused persons could be forced to choose between compromising their fair trial rights by proceeding with their bail application without a publication ban, or foregoing the right to seek bail in order to ensure that prejudicial information disclosed in the course of the bail hearing is not published.

[202] Because the order provided for under s. 517 is mandatory, it is rationally connected to the objective of allowing the accused to receive an expeditious bail hearing and at the same time protect the accused's right to a fair trial with an impartial jury. [page147] Minimal Impairment (i) The publication ban itself

[203] Rosenberg J.A. concluded that s. 517 meets the minimal impairment test. I agree to the extent that it applies to matters that may be tried by a jury. As with the rational connection criterion, because the publication ban applies to bail hearings in respect of all charges and not just charges that have the potential to be heard by a jury, s. 517 cannot be said to minimally impair the infringement of s. 2(b) of the Charter. The Attorney General for Ontario has suggested that the law is reasonably tailored, despite its application where a jury trial is not possible, because it protects the accused on occasions when an accused is initially charged with a summary conviction offence and is later charged with more serious offences that could potentially be tried by a jury. However, if on any of these occasional circumstances there has been reporting of prejudicial information from the summary conviction bail hearing, and the accused has concerns regarding the ability to obtain a fair trial before a jury on the later, more serious charges as a result, steps can be taken before trial to curtail the prejudice, such as, for example, challenge for cause or change of venue. (ii) The mandatory nature of the publication ban

[204] Are there alternatives to a mandatory publication ban that would be as effective to protect an accused from a potentially tainted jury pool but would have a more limited impact on curtailing freedom of expression and freedom of the press? Even if there are, did Parliament make a reasonable choice among available alternatives in providing a mandatory publication ban on bail proceedings? McLachlin J. explained the proper approach to the issue of minimal impairment in RJR- MacDonald Inc., at para. 160:

The impairment must be "minimal", that is, the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement . . . On the other hand, if the government fails to explain why a significantly less intrusive and equally effective measure was not chosen, the law may fail.

[205] I have already discussed why a discretionary ban would be both a procedurally and substantively ineffective remedy for an accused person. [page148]

[206] In his discussion of minimal impairment, Rosenberg J.A. outlines other alternative measures that could minimize the effect of a publication ban on freedom of the press, together with his reasons for rejecting each as an alternative that would not adequately achieve the fair trial objectives of s. 517. I agree with his discussion and conclusions.

[207] One of the alternative measures that Rosenberg J.A. considers and initially rejects at the minimal impairment stage of his analysis is a discretionary, tailored ban limited to specific types of prejudicial information that would be inadmissible at trial, such as an involuntary confession, bad character evidence, similar fact evidence or an accused's full criminal record.

[208] Another alternative that should also be considered is a mandatory ban that is limited to a specified list of such evidence or information. Such a ban has all the benefits of being mandatory: no delay of the bail hearing, no uncertainty for the accused that may cause the accused to have to choose between bail and the right to a fair trial, no extra time demands on the bail proceeding, no difficulties for the accused in preparing for and conducting a hearing to meet the Dagenais test. By being limited to information that is clearly prejudicial and possibly inadmissible at the trial, such a ban would, in theory, have a much more limited effect on freedom of the press by allowing the reporting of some information from the bail hearing, including all or some of the reasons for granting or denying bail.

[209] However, a mandatory ban that is limited to specific types of evidence would not be "equally effective", to use the language of RJR MacDonald, at para. 160. One problem is that it could be very difficult for Parliament to draft a sufficiently comprehensive list of banned prejudicial information. More importantly, as a practical matter, the press may find it unworkable to be circumscribed in what it is able to report from the bail hearing, especially if it is the banned, prejudicial evidence that is critical to the decision of the justice.

[210] In Canada (Attorney General) v. JTI-Macdonald Corp., 2007 SCC 30, [2007] 2 S.C.R. 610, [2007] S.C.J. No. 30, at para. 43, McLachlin C.J.C., writing for the court, emphasized the importance of deference by the courts where Parliament seeks to address multi-faceted problems:

Again, a certain measure of deference may be appropriate, where the problem Parliament is tackling is a complex social problem. There may be many ways to approach a particular problem, and no certainty as to which will be the most effective. It may, in the calm of the courtroom, be possible [page149] to imagine a solution that impairs the right at stake less than the solution Parliament has adopted. But one must also ask whether the alternative would be reasonably effective when weighed against the means chosen by Parliament. To complicate matters, a particular legislative regime may have a number of goals, and impairing a right minimally in the furtherance of one particular goal may inhibit achieving another goal. Crafting legislative solutions to complex problems is necessarily a complex task. It is a task that requires weighing and balancing. For this reason, this court has held that on complex social issues, the minimal impairment requirement is met if Parliament has chosen one of several reasonable alternatives: R. v. Edwards Books and Art Ltd., 1986 12 (SCC), [1986] 2 S.C.R. 713, [1986] S.C.J. No. 70; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927, [1989] S.C.J. No. 36.

[211] In this case, the legislative approach chosen by Parliament represents a reasonable alternative which effectively protects an accused's fair trial rights. It is the role of the court to respect Parliament's choice unless it does not pass constitutional muster. Proportionality between Salutary and Deleterious Effects

[212] Rosenberg J.A. found that the legislation fails at this stage of the s. 1 analysis because in his view, the deleterious effects of the legislation outweigh its salutary effects. He found that the deleterious effects are substantial while the salutary effects are speculative and limited. It is at this stage of the analysis where we have our major disagreement.

[213] The requirements of the third branch of the Oakes proportionality test were articulated by Lamer C.J.C. in Dagenais, where the context was an in-depth consideration of the constitutional parameters to be considered when a court is asked to order a publication ban at common law. There must be proportionality between the objective of a measure and its deleterious effects, and also between the actual salutary effects of the measure and its deleterious effects. Lamer C.J.C. explained how this proportionality analysis applies when deciding whether to order a common-law publication ban, at pp. 889-90 S.C.R.:

A similar view of proportionality must inform the common law rule governing publication bans . . . This suggests that when a ban has a serious deleterious effect on freedom of expression and has few salutary effects on the fairness of a trial, the ban will not be authorized at common law.

It is also important to recognize, however, that the objective usually underlying such bans -- the diminution of the risk that a trial might be tainted by unfairness -- is directly related to the accused's constitutionally protected right to a fair trial. Although, as I noted earlier [citation omitted], it is incorrect to oversimplify the relationship between the right to a fair trial and the right to freedom of expression by blindly applying the "clash of rights" model, there are times when the rights of the accused [page150] will be in direct conflict with the expressive rights of the media. In such cases, I believe it is necessary to apply the common law proportionality analysis in a manner that reflects the fact that two fundamental rights are in jeopardy. That is, it is essential in these circumstances to recognize that the pressing and substantial objective at issue is itself a fundamental right, and that, as such, it is a matter of exceptional importance. This will be of particular significance when considering whether there are reasonable alternative measures available, and when assessing the balance between the salutary and deleterious effects of the ban. When examining alternative measures, it will be important to carefully consider both rights at issue, so as to ensure that any alternative measures that impair free expression to a lesser degree than a publication ban also reasonably protect the right to a fair trial. Similarly, when considering the proportionality of the impact of the ban on free expression to its salutary effects on the fairness of the trial, it will be necessary to bear in mind the fundamental importance of trial fairness, both to the accused and to society. (Emphasis in original)

[214] In his reasons, Lamer C.J.C. rejected the notion that juries are always influenced by publications, but recognized as a matter of common sense that some jurors will be adversely affected by what they read, see or hear. He expressed confidence that jurors can follow judicial instructions to ignore pre-trial information, but also expressed concern that this will be more difficult when the publicity is not from a finite source but sustained over time. Such sustained pre-trial publicity could become implanted into the minds of jurors to the extent that they become unable to distinguish the pre-trial information from the evidence at trial.

[215] As I have noted, Lamer C.J.C. conceded that it will be a rare case when a publication ban will have no influence on the fairness of the trial. Therefore, he did not discount the potential for juror contamination and for trial unfairness because of prejudicial pre-trial publicity. Instead, he concluded that when considering an accused's request for a common-law publication ban on pre-trial publicity about the case, a court should assess whether there are alternative measures to accomplish the same purpose.

[216] With that background in mind, I turn to the salutary and deleterious effects of a publication ban on bail proceedings. (i) Deleterious effects of the publication ban itself identified by Rosenberg J.A.

[217] Rosenberg J.A. lists seven deleterious effects of the mandatory ban. Two relate directly to the ban itself: (1) the timing of the ban stifles discussion about the bail hearing and the reasons why bail was granted or denied both at the time of the decision or [page151] later if, for example, bail is granted and later revoked. Although the ban is temporary, the issue will not be as newsworthy after trial; and, (2) the ban is unfair to a co-accused who may want the pre-trial publicity. (ii) Deleterious effects because the ban is mandatory at the request of the accused identified by Rosenberg J.A.

[218] The other five deleterious effects relate to the fact that the ban is mandatory. The first is that the ban extends to charges which can only be tried by a judge alone. I have already stated that the mandatory ban is overly broad in this regard and cannot be justified under s. 1. In the remedy I will propose, both the Crown and the accused may still seek a discretionary ban in non-jury cases on grounds other than trial fairness based on the potential for jury contamination.

[219] The other four deleterious effects identified by Rosenberg J.A. are: (1) the ban extends to all information, including non-prejudicial information, and to information whose potential prejudice could be addressed by alternative measures such as challenge for cause, change of venue and strong judicial direction; (2) the ban prohibits publication of the reasons of the justice for granting or denying bail, even when the reasons do not interfere with the accused's fair trial rights; (3) the ban operates where prejudicial information is already in the public realm, as in these cases where police spoke to the media at the time they arrested the accused; and (4) the ban will prevent individuals with knowledge of the case from hearing about it and coming forward with relevant information in a timely way.

[220] Although I have identified these effects as arising from the fact the ban is mandatory, it would be more accurate to say that they arise from the fact that it is a mandatory complete ban on all information. As I discussed in the minimal impairment section, the ban could have been enacted as a mandatory ban only applicable to a list of categories of prejudicial information such as confessions, in which case some or even all of these deleterious effects would not occur. However, Parliament chose to impose a general ban. The issue at this stage of the analysis is whether that choice causes the legislation to fail the Oakes test if these deleterious effects are not outweighed by the salutary effect of a full mandatory ban.

[221] Before moving to the salutary effects of the mandatory ban, I wish to comment on a number of the issues raised by Rosenberg J.A. in his discussion of deleterious effects. The first is the notion that alternative measures such as challenge for cause, [page152] change of venue or strong jury instructions could eliminate the risk to a fair trial from publication of prejudicial information from the bail hearing. There are two serious practical problems with this suggestion. First, as I discussed earlier, these alternative orders are not available to a justice on a bail hearing. As well, it would be impractical and costly for accused persons, the legal aid system and the justice system to change the venues of jury trials on a regular basis as a method of countering the adverse effect on a jury of pre-trial prejudicial information from a bail hearing. [See Note 11 below] Second, a change of venue may no longer be an effective remedy in any event, because the Internet preserves pre-trial publicity and search engines like Google make it easy for any person to access this information, regardless of location or time.

[222] The second observation I have relates to situations where there has already been pre-trial publicity from sources such as a police press conference. I reject the argument that in such situations there is no longer any need for a publication ban on similar prejudicial information that may come out at the bail hearing. I reject this argument for two reasons. The first is that the police must also be concerned with the fair trial rights of accused persons when they make information public outside a court setting. The fact that police disclosure has not been challenged in court does not mean that there are no limits on that activity. In that context, it is not useful for the court to use the fact that the police choose to publish some information they have about an accused as a basis for allowing the press to report the same or other potentially prejudicial information from the bail hearing.

[223] Second, and more importantly, there is an enormous difference, both in fact and in public perception, between information that is spoken outside court and information revealed in court proceedings. Although the rules of evidence are not strictly applied in bail hearings, a bail hearing is a court hearing and a report of evidence from a court hearing will be treated seriously by the public. By contrast, the public knows that information disclosed at a press conference is untested and it will be treated with much more scepticism by an informed public. The same can be said about an informed public's reaction to local gossip: see Rosenberg J.A.'s reasons at para. 89, quoting the Ontario Royal Commission Inquiry into [page153] Civil Rights, Report Number One, vol. 2 (Ottawa: Queen's Printer, 1968), at p. 763.

[224] Finally, I do not agree that a publication ban on bail proceedings will necessarily prevent witnesses from coming forward. If that is a real concern, there are ways to address it, such as carefully conducted police press conferences. I also do not agree that a co-accused is significantly prejudiced by such a ban. The co-accuseds' complaint appears to be with pre-trial publicity by the police and not with the courts or the administration of justice by the courts. (iii) Salutary effects of the publication ban itself

[225] The major salutary effect of the mandatory publication ban on bail proceedings is that it protects the fair trial rights of an accused from being undermined by a jury that may have been tainted by prejudicial information that will not be admissible at trial. Rosenberg J.A. views this salutary effect with scepticism, which is the main reason why he considers the deleterious effects as overriding. He views the social science research as inconclusive on the effect of pre-trial publicity on a jury.

[226] However, the Supreme Court has made clear in Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, [2004] S.C.J. No. 28, at para. 77, that in order to justify a provision targeted to address a harm that is not measurable, the government is not required to provide scientific proof of the harm. Rather"[w]here the court is faced with inconclusive or competing social science evidence relating the harm to the legislature's measures, the court may rely on a reasoned apprehension of that harm". This principle was recently applied by Sharpe J.A. in Cochrane v. Ontario (Attorney General) (2008), 2008 ONCA 718, 92 O.R. (3d) 321, [2008] O.J. No. 4165 (C.A.).

[227] The Supreme Court has acknowledged both in Corbett and in Dagenais that, although the effect of inadmissible information on a jury is not measurable, and although we place great trust in our juries, as a matter of logic and common sense, some jurors will be adversely affected by the knowledge of prejudicial information that is not part of the evidence at trial. It seems to me that it would be contrary to the decisions in these cases, where the issue was thoroughly considered (also see the dissenting reasons of La Forest J. in Corbett) and the link between pre-trial publication and potential jury contamination was accepted, to strike down legislation as unconstitutional on the basis that that link is speculative. [page154]

[228] While accepting that some particularly prejudicial inadmissible information may influence jurors, Rosenberg J.A. rejects the position that a publication ban has a salutary effect and suggests instead a more targeted remedial order, such as a ban closer to trial or a change of venue as the better answer. However, it is clear that a ban on publication of the bail proceeding will have a salutary effect on fair trial rights in such a case. In contrast, for reasons I have already stated, the remedial measures suggested, while available in theory, would be costly for the accused and for the administration of justice both to obtain and to implement.

[229] At this point in time, it is unclear how legal aid will approach and address the problem that counsel will require funding to seek discretionary publication bans or tailored remedial orders such as change of venue, as well as to conduct trials out of town if changes of venue are ordered. We know that legal aid funds are limited and are being stretched to accommodate increasing needs. This is also a practical factor when we are discussing the salutary versus deleterious effects of a current measure that is simply applied and cost-effective.

[230] Rosenberg J.A. refers to and essentially discounts the following four incidental salutary effects of a mandatory publication ban of bail proceedings: prevention of witness contamination, protection of ongoing criminal investigations, encouragement of witnesses and sureties to come forward and advancement of the privacy interests of the accused. While these may be viewed as extra benefits of a ban, I would give them no weight at this stage of the analysis. In my view, it was not the intent of Lamer C.J.C. in requiring a salutary versus deleterious effects comparison, that the court be overly concerned with the extent to which any side benefits of the measure may or may not be effective. The question in this case is whether, in practice, the deleterious effects are sufficient to overcome the benefit of an important measure, intended to protect the fair trial rights of accused persons under our law.

[231] I would add as a salutary effect the fact that, under s. 517, the ban is time limited and does not preclude the public and the press from attending the hearing. The process is open, and once the trial is complete, there is the opportunity for full discussion of any information that was not available to be reported earlier. I agree that any limit on the ability of the press to report or on the timing of when it can report on court proceedings is a serious breach of s. 2(b). However, the temporary nature of the ban ensures that the proceedings are not in any way being hidden from the public or that they will not be [page155] subject to public scrutiny. Of course, the Code also contains a so far unchallenged [See Note 12 below] similar mandatory ban at the request of the accused on preliminary inquiry evidence in s. 539. As well, regardless of whether the accused requests the preliminary inquiry publication ban, s. 542(2) prohibits anyone from publishing reports of an admission or confession given at a preliminary inquiry until the end of the trial. All these provisions work together and are timed to operate together to achieve the same purpose.

[232] The appellants argue that there are only currently approximately 500 jury trials per year in Ontario and that there has never been a problem finding an impartial jury. However, these cases often involve the most serious charges and are therefore cases where the media may be most interested in reporting on the bail proceedings. [See Note 13 below] It is logical to conclude that the current provisions have contributed to preserving the fair trial rights of those accused. (iv) Salutary effects because the ban is mandatory at the request of the accused and not discretionary

[233] Rosenberg J.A. identifies avoidance of delay and expense as the most significant salutary effect of a mandatory, as opposed to a discretionary, ban. His solution for this problem in the context of a discretionary ban regime is his decision that a justice is not required to give notice to the media when an accused asks for a discretionary ban to be ordered. However, avoidance of delay and expense for the accused is just one of the salutary effects of the ban being mandatory and not discretionary.

[234] Another salutary effect is the length of time it would take to hold a discretionary ban hearing in each case. As well, adjournments will be needed for the system to try to accommodate lengthy bail hearings. As I discussed earlier, whether or not [page156] the media are present and opposing the ban, the onus will be on the accused to justify the ban and to meet the Dagenais test.

[235] Even more important is the unlikelihood that an accused would be able to meet the Dagenais test for obtaining a discretionary ban in an expeditious fashion or at all at this very early stage of a criminal proceeding. Given that an accused seeking bail has generally not had the opportunity of either retaining counsel or obtaining Crown disclosure, it may simply be unrealistic to expect the accused to be able to satisfy the Dagenais test in an expeditious fashion, if at all, at this stage of the proceeding. [See Note 14 below] Conclusion on Proportional Effects

[236] In Canada (Attorney General) v. JTI-Macdonald Corp., supra, McLachlin C.J.C. described the last part of the proportionality test as follows, at para. 45:

The final question is whether there is proportionality between the effects of the measure that limits the right and the law's objective. This inquiry focuses on the practical impact of the law. What benefits will the measure yield in terms of the collective good sought to be achieved? How important is the limitation on the right? When one is weighed against the other, is the limitation justified? (Emphasis in original)

[237] In other words, after considering the strength of the law's objective, the rational connection of the law to the objective and whether the law minimally impairs the Charter right that is being violated, the court is still directed to stand back and ensure that the positive effects of the law justify its negative impact on the affected Charter right.

[238] In applying the test in the context of a publication ban sought to preserve fair trial rights, Lamer C.J.C. in Dagenais, at pp. 889-90 S.C.R., also directed that when the objective of the ban is to protect another Charter right, then two fundamental rights are in jeopardy and "it is essential in these circumstances to recognize that the pressing and substantial objective at issue is itself a fundamental right, and that, as such, it is a matter of exceptional importance".

[239] In this case, the positive effect of the mandatory publication ban is to protect the fair trial rights of the accused by ensuring that an accused seeking bail on a charge where a jury trial is possible will know that if asked for, the court will order a [page157] publication ban on all information disclosed at the bail hearing, including prejudicial information that may not be admissible at the trial. The accused need not be concerned about seeking bail and about what information may be disclosed at the hearing because the accused is guaranteed that any future jurors will not hear reports of potentially prejudicial information from the bail hearing which could affect their verdict and, therefore, the accused's fair trial.

[240] The fact that the publication ban is mandatory and not discretionary is one of its most significant benefits. The Dagenais common-law publication ban process is ill-suited to the bail context and structure, both procedurally and substantively. If only a discretionary ban were available following a Dagenais hearing, the accused's right to reasonable bail, including an expeditious and cost-effective process, could be undermined. In Canadian Newspapers Co. v. Canada (Attorney General), 1988 52 (SCC), [1988] 2 S.C.R. 122, [1988] S.C.J. No. 67, where the issue was the mandatory nature of a publication ban on the identity of sexual assault victims, the Supreme Court recognized that the ban had to be mandatory, otherwise victims might not come forward out of fear that they would be publicly identified. Similarly, an accused needs to be certain when seeking bail that whatever prejudicial information may be disclosed at the bail hearing will be protected from disclosure to potential jurors. If the ban is not mandatory, then the accused's right to reasonable bail and, at the same time, the accused's liberty interest may be undermined.

[241] Requiring accused to regularly seek discretionary bans would also detrimentally affect the functioning of the bail system in Ontario, which, certainly in some regions, does not appear to be equipped to handle extended hearings and still deal with bail requests expeditiously.

[242] As I have already stated in connection with the publication ban itself, it is also a salutary feature of the mandatory ban that it is not permanent, nor does it involve an in camera hearing. The ban is only temporary until the jury trial is complete. The press and the public are allowed to attend as with any court proceeding. Rosenberg J.A. recognizes these features in his analysis but does not give them weight as salutary effects. In my view, they are significant features at this stage of the proportionality analysis because they have the effect of minimizing the limitation on freedom of the press.

[243] It is important to note that Parliament amended and re- enacted s. 517 in 2005 to include publication "in any document". It did this after the Supreme Court's decision in Dagenais. [page158] In so doing, Parliament can be seen to have rejected the recommendation of the Law Reform Commission of Canada in Working Paper 56, Public Media Access to the Criminal Process (Ottawa, 1987) that the mandatory publication ban be repealed because of its effect on freedom of expression, and as a reaffirmation by Parliament of the choice it made in enacting s. 517 to balance the fair trial right with freedom of expression in the limited circumstance of a bail hearing.

[244] The deleterious effect of the publication ban is its limit on freedom of expression and freedom of the press by preventing the publication of any information from the bail hearing, including the reasons for granting or denying bail. These Charter freedoms are critical for a free and democratic society and their importance cannot be minimized. That said, as discussed above, the limit is temporary and the court remains open.

[245] I do not consider the fact that the ban is mandatory at the request of the accused to be a deleterious effect for the reasons I have already discussed. In my view, it is essential for the protection of the accused, for the operation of the bail system and for the efficient and effective administration of justice in Ontario that routine requests for a publication ban on bail proceedings where there may be a jury trial result in a mandatory order, not a discretionary hearing procedure.

[246] I acknowledge that a deleterious effect of the legislation is that it extends to all information at the bail hearing and it is not tailored to cover only the information that is necessary to protect the fair trial rights of the accused. However, it is unclear that a more tailored ban would make a significant difference to the press. In cases where the press wish to report on bail proceedings, they will want to report fully and fairly and not be constrained to leave out potentially prejudicial material, such as a confession, criminal record, or similar fact evidence, especially where the judge's decision regarding bail turns to any extent on that material. This practical concern may well explain Parliament's decision to enact a full mandatory ban.

[247] In my view, the legislation reflects a reasonable choice by Parliament to enact a measure that meets its objective to promote the Charter fair trial right of the accused and is minimally impairing of the Charter right to freedom of expression. I conclude that, weighing the relevant salutary and deleterious effects of the legislation, the salutary effects outweigh the deleterious effects and, therefore, the limitation on freedom of expression is justified. [page159] Remedy

[248] I have concluded that the limit on freedom of expression in s. 517 is justified to prevent the potential breach of the fair trial rights of an accused whose charges may possibly be tried by a jury. However, to the extent that the mandatory ban on bail proceedings applies in respect of charges that cannot be tried by a jury, the legislation is not justified under s. 1.

[249] I agree with Rosenberg J.A. that reading down the legislation is the appropriate remedy for this legislation. In order to read down the section so that it will only apply where the accused may be tried by a jury, some words must be added. This was the remedy applied in R. v. Sharpe, 2001 SCC 2, [2001] 1 S.C.R. 45, [2001] S.C.J. No. 3, at para. 114, where, in order to cure a problem of overbreadth, McLachlin C.J.C. held that the appropriate remedy was "to read into the law an exclusion of the problematic application of s. 163.1". A similar approach was taken in Ruby v. Canada (Solicitor General), 2002 SCC 75, [2002] 4 S.C.R. 3, [2002] S.C.J. No. 73, at para. 60, where Arbour J. read down s. 51(2)(a) of the Privacy Act, R.S.C. 1985, c. P-21, by adding language to limit in camera hearings only to those circumstances where the government institution chose to make ex parte representations.

[250] I propose the following revision to s. 517 in order to read down the mandatory ban:

517(1) If the prosecutor or the accused intends to show cause under section 515, he or she shall so state to the justice and the justice may, and shall on application by the accused where and for so long as the charge(s) may be tried by a jury, ... (Emphasis added)

[251] The effect of the added phrase is to limit the mandatory publication ban at the request of the accused to bail proceedings in respect of charges that can possibly be tried by a court composed of a judge and jury. Consistent with the goal of effectively addressing the overbreadth problem while ensuring that an accused's fair trial rights are preserved, the duration of the ban is limited, so that if through an election by the Crown or the defence the trial becomes a judge alone trial, either in the Ontario Court or the Superior Court, the ban no longer applies.

[252] I realize that this revision to the duration of the ban does not co-ordinate precisely with the similar mandatory publication ban on preliminary inquiry evidence under s. 539 of the Code, where the ban continues in force until the accused is discharged or until the conclusion of the trial, whether by jury or [page160] judge alone. However, this approach is most consistent with the analysis that justifies the full mandatory ban on bail proceedings in order to ensure the fair trial of the accused before a jury. Parliament may well wish to consider and address both provisions and to apply its own wording to achieve the constitutional objectives of the legislation.

[253] There is no need to suspend the operation of the section to accommodate the proposed read-down. The section will continue to operate as before, except it will apply in fewer cases. The Availability of Discretionary Bans

[254] This formulation still allows an accused who is charged with an offence that cannot be tried by a jury to seek a discretionary publication ban of the bail proceedings. In R. v. Mentuck, 2001 SCC 76, [2001] 3 S.C.R. 442, [2001] S.C.J. No. 73, 158 C.C.C. (3d) 449, at para. 32, Iacobucci J. explained that a publication ban can be sought to protect "other crucial aspects of the administration of justice". For example, if an accused will not be able to obtain bail without sureties and the court is satisfied that the identity of the sureties should be protected, a discretionary ban could be ordered to ensure that the accused is able to exercise the right to obtain reasonable bail. Presumably, such orders will be relatively rare and therefore any delays caused by such orders will not be so extensive as to affect the justice system as a whole. Disposition

[255] For these reasons, I would allow the appeal from the order of Durno J. dated March 1, 2007 in part. I would read down s. 517 by adding the words "where and for so long as the charge(s) may be tried by a jury" after "shall on application by the accused". Otherwise, I would dismiss the appeal.

[256] I agree with Rosenberg J.A. that this is not a case for costs.

Appeal allowed in part.

Notes

Note 1: There were minor amendments made to the provision in An Act to amend the Criminal Code, S.C. 2005, c. 32. Changes were made to the wording - for example, the word "newspaper" was replaced with "document" and the phrase "or transmitted in any way" was added after the word "broadcast".

Note 2: Interestingly, not long after writing his reasons in Global Communications, Linden J. had second thoughts about the decision, since the trial of the fugitive was to take place in the United States, which has extensive safeguards to ensure a fair trial despite pre-trial publicity: see Allen M. Linden"Limitations on Media Coverage of Legal Proceedings: A Critique and Some Proposals for Reform" in Philip Anisman and Allen M. Linden, eds., The Media, The Courts and the Charter (Toronto: Carswell, 1986) 301, at p. 319.

Note 3: One such circumstance is where the accused testifies and his record is introduced under s. 12 of the Canada Evidence Act, R.S.C. 1985, c. C-5.

Note 4: Ninety-five per cent of criminal cases are tried in the Ontario Court of Justice: see the Ontario Court of Justice 2005 Annual Report, online: http:. Of the cases that proceed to the Superior Court of Justice, slightly less than half are tried by jury.

Note 5: See the 2006 reasons of the application judge, at paras. 120-21, and R. v. Pearson, 1992 52 (SCC), [1992] 3 S.C.R. 665, [1992] S.C.J. No. 99, 77 C.C.C. (3d) 124, at pp. 682-85 S.C.R., pp. 135-36 C.C.C.

Note 6: Kerri L. Pickel"Inducing Jurors to Disregard Inadmissible Evidence: A Legal Explanation Does Not Help" (1995), 19 Law and Human Behavior 407; Edith Greene and Mary Dodge"The Influence of Prior Record Evidence on Juror Decision Making" (1995), 19 Law and Human Behavior 67.

Note 7: Kramer et al., at p. 414.

Note 8: R. v. Brooks, 2001 28401 (ON SC), [2001] O.J. No. 1563, 153 C.C.C. (3d) 533 (S.C.J.), at paras. 34-39.

Note 9: See Lisa Dufraimont"Evidence Law and the Jury: A Reassessment" (2008), 53 McGill L.J. 199, at 218, where Prof. Dufraimont notes that "modern trial procedure incorporates a complex system of jury-control devices, including jury-selection rules aimed at excluding biased individuals, evidentiary rules that control the flow of information at trial, binding judicial instructions on the law, and influential judicial comments on the evidence".

Note 10: These concerns would not apply on the infrequent occasions when a discretionary ban is being sought for other reasons, such as in cases where the ban is sought on application by the Crown.

Note 11: According to Ontario Court Input Statistical System data, in 2005-2006, there were 487 jury trials in Ontario.

Note 12: Section 539 was upheld as a minimal infringement of s. 2(b) in R. v. Banville, 1983 3027 (NB QB), [1983] N.B.J. No. 110, 3 C.C.C. (3d) 312 (Q.B.). However, that case was decided before Dagenais.

Note 13: See Lisa Dufraimont"Evidence Law and the Jury: A Reassessment", supra, note 9, at 209-210. Although Prof. Dufraimont notes that criminal jury trials are rare exceptions when compared against the total number of criminal cases in Canada, she further acknowledges that "jury trials nonetheless retain a systemic importance that belies their numbers because many of the most serious and important criminal cases are tried by juries. The gravest offences in Canada's Criminal Code, notably murder and treason, are normally required to be tried by juries."

Note 14: The discretionary ban is postulated as what is left if the mandatory aspect of s. 517 is struck down as unconstitutional, based on Rosenberg J.A.'s analysis.