The Attorney General of Canada on Behalf of the Republic of France et al. v. Diab

[Indexed as: France v. Diab]

Ontario Reports

Court of Appeal for Ontario,

Hoy A.C.J.O., Blair and Rouleau JJ.A.

May 15, 2014

120 O.R. (3d) 174 | 2014 ONCA 374

Case Summary

Charter of Rights and Freedoms — Fundamental justice — Extradition — Person facing surrender who alleges that case against him relies on evidence derived from torture required to establish plausible connection between challenged evidence and use of torture — If that threshold is met, minister required to make further inquiries to determine whether there is real risk that torture-derived evidence will be used in proposed foreign proceeding — Minister generally required to refuse surrender if he concludes that real risk exists — Torture and torture-derived evidence undermining right to fair trial — Use of torture and torture-derived evidence violating principles of fundamental justice.

Criminal law — Extradition — Appeal — Extradition judge finding that French handwriting analysis that linked appellant to suspected bombing was not manifestly unreliable and tipped scale in favour of committal — Extradition judge wondering whether there was unique French methodology with respect to handwriting comparison analysis that explained markedly divergent opinions between French experts and appellant's experts — Extradition judge refusing appellant's request for adjournment to permit him to obtain further evidence with respect to French methodology on basis that proposed evidence would not affect his decision — Appellant's application to adduce fresh evidence on appeal to establish that there is no unique French methodology dismissed — Proposed fresh evidence could not reasonably have affected result.

Criminal law — Extradition — Committal — France seeking appellant's extradition for his alleged role in 1980 bombing outside Paris synagogue — Extradition judge finding that four of five pieces of evidence in record of the case were insufficient alone to justify committal but that handwriting analysis that linked appellant to suspected bombing was not manifestly unreliable and tipped scale in favour of committal — Appellant's appeal dismissed — Threshold reliability test for admission of expert evidence at trial not applicable in extradition context — Extradition judge's conclusion that report was not manifestly unreliable [page175] not rooted in impermissible speculation or misappropriation of evidence — Having determined that report was not manifestly unreliable, extradition judge not erring in failing to analyze report further to decide if it would be dangerous or unsafe to convict on all evidence in record of the case.

Extradition — Surrender — Jurisdiction — Minister having jurisdiction to order appellant's surrender even though requesting state had not yet decided whether to put appellant on trial — Requirement in s. 3(1) of Extradition Act that extradition request be for "purpose of prosecuting the person" satisfied if process has been initiated that will lead to trial if not discontinued — Trial not having to be inevitable — Extradition Act, S.C. 1999, c. 18, s. 3(1).

France sought the appellant's extradition as a result of his alleged role in a 1980 bombing outside a Paris synagogue. The extradition judge found that four of the five pieces of evidence in the record of the case were insufficient alone to justify committing the appellant for extradition. The remaining piece of evidence, a French handwriting analysis that linked the appellant to the suspected bomber, was dismissed by the appellant's experts as methodologically flawed and unreliable. The extradition judge agreed that the French expert's conclusions were "suspect", but concluded that her evidence could not be completely rejected as manifestly unreliable and that it tipped the scale in favour of committal. The extradition judge wondered whether there was a unique French methodology with respect to handwriting analysis that explained the markedly divergent opinions between the French experts and the appellant's experts. The appellant sought an adjournment to permit him to obtain further evidence with respect to French methodology. The extradition judge refused the request on the ground that the proposed evidence would not affect his decision. The appellant was ordered committed for extradition. Before the Minister of Justice, the appellant argued that the minister lacked jurisdiction to surrender him to France because France had not yet made a decision that he would stand trial. The minister rejected that submission. The appellant also argued that it would be unjust, oppressive and contrary to the Canadian Charter of Rights and Freedoms to extradite him on the basis of evidence that was plausibly connected to the use of torture. The minister noted that, as a party to the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, France has a duty to ensure that evidence derived from torture is not used in any legal proceedings. The minister ordered the appellant's surrender. The appellant appealed the committal order and brought an application for judicial review of the surrender order. He sought to adduce fresh evidence on appeal to establish that there is no unique French methodology for handwriting analysis.

Held, the appeal and the application should be dismissed.

The proposed fresh evidence did not meet the test for admission as it could not reasonably be expected to have affected the result.

The threshold reliability test for the admission of expert evidence at trial has no application in the extradition context. The test for committal is made out where (1) evidence that is manifestly unreliable, and on which it would therefore be dangerous or unsafe to convict, is excluded; and (2) there remains evidence available for trial that possesses indicia of threshold reliability on each essential element of the corresponding Canadian offence, upon which a reasonable jury properly instructed could convict. The extradition judge's conclusion that the handwriting expert's report was not manifestly unreliable was not rooted in impermissible speculation or a misapprehension of the evidence. Having found [page176] that the report was not manifestly unreliable, the extradition judge did not err by not analyzing the evidence further to decide if it would be dangerous or unsafe to convict on all of the evidence in the record of the case. The "dangerous or unsafe to convict" standard properly belongs in the test for manifest unreliability, where the extradition judge applied it.

The Extradition Act does not limit the minister's jurisdiction to order extradition to cases where the requesting state has decided to put the person on trial. The requirement in s. 3(1) of the Act that the extradition request be for the "purpose of prosecuting the person" is satisfied if a process has been initiated that, if not discontinued, will lead to trial. The person sought must be more than a mere suspect. A trial of that person, however, need not be inevitable. The minister reasonably concluded that the French authorities had taken steps consistent with the commencement of a prosecution against the appellant.

Torture and torture-derived evidence undermine the right to a fair trial, and their use therefore violates the principles of fundamental justice contrary to s. 7 of the Charter. A person facing extradition who alleges that the case against him relies on evidence derived from torture is required to establish a plausible connection between the challenged evidence and the use of torture. If that threshold is met, the minister is required to make further inquiries to determine whether there is a real risk that torture-derived evidence will be used in the proposed foreign proceeding. The minister is generally required to refuse surrender if he concludes that a real risk exists. In this case, the minister may have conflated the two steps of the inquiry and did not expressly state that he was satisfied that there was no real risk that torture-derived evidence would be used, but it was apparent from the conclusions he reached that he was indeed so satisfied.

Dunsmuir v. New Brunswick, [2008] 1 S.C.R. 190, [2008] S.C.J. No. 9, 2008 SCC 9, 329 N.B.R. (2d) 1, 64 C.C.E.L. (3d) 1, EYB 2008-130674, J.E. 2008-547, [2008] CLLC Â220-020, 170 L.A.C. (4th) 1, 372 N.R. 1, 69 Imm. L.R. (3d) 1, 291 D.L.R. (4th) 577, 69 Admin. L.R. (4th) 1, 95 L.C.R. 65, D.T.E. 2008T-223, 164 A.C.W.S. (3d) 727; R. v. Palmer, 1979 8 (SCC), [1980] 1 S.C.R. 759, [1979] S.C.J. No. 126, 106 D.L.R. (3d) 212, 30 N.R. 181, 50 C.C.C. (2d) 193, 14 C.R. (3d) 22, 17 C.R. (3d) 34, 4 W.C.B. 171, apld

Alberta (Information and Privacy Commissioner) v. Alberta Teachers' Assn., [2011] 3 S.C.R. 654, [2011] S.C.J. No. 61, 2011 SCC 61, 2011EXP-3798, J.E. 2011-2083, 424 N.R. 70, 339 D.L.R. (4th) 428, 28 Admin. L.R. (5th) 177, 52 Alta. L.R. (5th) 1, [2012] 2 W.W.R. 434, 519 A.R. 1, 208 A.C.W.S. (3d) 434; Fletcher v. Government of France (2008), 2007-09 Gib. L.R. 191 (S.C. Gibraltar); India v. Singh, 1996 17926 (BC SC), [1996] B.C.J. No. 2792, 108 C.C.C. (3d) 274, 36 Imm. L.R. (2d) 17, 31 W.C.B. (2d) 428 (S.C.); Li v. Canada (Minister of Citizenship and Immigration), [2005] F.C.J. No. 1, 2005 FCA 1, [2005] 3 F.C.R. 239, 249 D.L.R. (4th) 306, 329 N.R. 346, 41 Imm. L.R. (3d) 157, 136 A.C.W.S. (3d) 345, affg [2003] F.C.J. No. 1934, 2003 FC 1514, [2004] 3 F.C.R. 501, 243 F.T.R. 261, 39 Imm. L.R. (3d) 216, 128 A.C.W.S. (3d) 138 [Leave to appeal to S.C.C. refused [2005] S.C.C.A. No. 119]; R. v. Abbey (2009), 97 O.R. (3d) 330, [2009] O.J. No. 3534, 2009 ONCA 624, 68 C.R. (6th) 201, 254 O.A.C. 9, 246 C.C.C. (3d) 301 [Leave to appeal to S.C.C. refused [2010] S.C.C.A. No. 125]; R. v. Mohan, 1994 80 (SCC), [1994] 2 S.C.R. 9, [1994] S.C.J. No. 36, 114 D.L.R. (4th) 419, 166 N.R. 245, J.E. 94-778, EYB 1994-67655, 71 O.A.C. 241, 89 C.C.C. (3d) 402, 29 C.R. (4th) 243, 23 W.C.B. (2d) 385; The Minister for Justice, Equality and Law Reform v. Bailey, [2012] IESC 16 (S.C. Ireland); United States of America v. Anderson (2007), 85 O.R. (3d) 380, [2007] O.J. No. 449, 2007 ONCA 84, 219 O.A.C. 369, 218 C.C.C. (3d) 225, 153 C.R.R. (2d) 20, 74 W.C.B. (2d) 353 [Leave to appeal to S.C.C. refused [2007] S.C.C.A. No. 159]; United States of America v. Ferras; United States of America v. Latty, [2006] 2 S.C.R. 77, [2006] S.C.J. No. 33, 2006 SCC 33, 268 D.L.R. (4th) 1, 351 N.R. 1, J.E. 2006-1461, 209 C.C.C. (3d) 353, 39 C.R. (6th) 207, 143 C.R.R. (2d) 140, EYB 2006-107828, 69 W.C.B. (2d) 711; United States of America v. Graham, [2007] B.C.J. No. 1390, 2007 BCCA 345, 243 B.C.A.C. 248, 222 C.C.C. (3d) 1, 74 W.C.B. (2d) 354 [Leave to appeal to S.C.C. refused [2007] S.C.C.A. No. 467]; [page177] United States of America v. Khadr (2011), 106 O.R. (3d) 449, [2011] O.J. No. 2060, 2011 ONCA 358, 234 C.R.R. (2d) 31, 280 O.A.C. 210, 85 C.R. (6th) 143, 273 C.C.C. (3d) 55, 337 D.L.R. (4th) 638 (C.A.), affg [2010] O.J. No. 3301, 2010 ONSC 4338, 215 C.R.R. (2d) 287, 258 C.C.C. (3d) 231, 78 C.R. (6th) 1, 322 D.L.R. (4th) 483 (S.C.J.) [Leave to appeal to S.C.C. refused [2011] S.C.C.A. No. 316]; United States of America v. Thomlison (2007), 84 O.R. (3d) 161, [2007] O.J. No. 246, 2007 ONCA 42, 219 O.A.C. 322, 68 Admin. L.R. (4th) 247, 216 C.C.C. (3d) 97, 72 W.C.B. (2d) 451 (C.A.) [Leave to appeal to S.C.C. refused [2007] S.C.C.A No. 179], consd

United States of Mexico v. Hurley (1997), 1997 3355 (ON CA), 35 O.R. (3d) 481, [1997] O.J. No. 2487, 101 O.A.C. 121, 116 C.C.C. (3d) 414, 8 C.R. (5th) 354, 45 C.R.R. (2d) 73, 35 W.C.B. (2d) 222 (C.A.), distd

Other cases referred to

A & Others v. Secretary of State for the Home Department (No. 2), [2005] UKHL 71, [2006] 2 A.C. 221 (H.L.); Arton (Re), [1896] 1 Q.B. 108 (Div. Ct.); Assange v. Swedish Prosecution Authority, [2012] UKSC 22, affg [2011] EWHC 2849 (Admin); Canada v. Schmidt, 1987 48 (SCC), [1987] 1 S.C.R. 500, [1987] S.C.J. No. 24, 39 D.L.R. (4th) 18, 76 N.R. 12, 20 O.A.C. 161, 33 C.C.C. (3d) 193, 58 C.R. (3d) 1, 28 C.R.R. 280, 2 W.C.B. (2d) 299; Canada (Citizenship and Immigration) v. Khosa, [2009] 1 S.C.R. 339, [2009] S.C.J. No. 12, 2009 SCC 12, 304 D.L.R. (4th) 1, 385 N.R. 206, 77 Imm. L.R. (3d) 1, 82 Admin. L.R. (4th) 1, EYB 2009-155418, J.E. 2009-481, 175 A.C.W.S. (3d) 7; Canada (Justice) v. Fischbacher, [2009] 3 S.C.R. 170, [2009] S.C.J. No. 46, 2009 SCC 46, 198 C.R.R. (2d) 168, 248 C.C.C. (3d) 419, EYB 2009-164941, J.E. 2009-1880, 394 N.R. 139, 69 C.R. (6th) 21, 255 O.A.C. 288, 312 D.L.R. (4th) 1; Canada (Minister of Justice) v. Pacificador (2002), 2002 41595 (ON CA), 60 O.R. (3d) 685, [2002] O.J. No. 3024, 216 D.L.R. (4th) 47, 162 O.A.C. 299, 166 C.C.C. (3d) 321, 6 C.R. (6th) 161, 97 C.R.R. (2d) 20, 55 W.C.B. (2d) 63 (C.A.) [Leave to appeal to S.C.C. refused [2002] S.C.C.A. No. 390, 101 C.R.R. (2d) 374]; Charkaoui (Re), [2004] F.C.J. No. 1236, 2004 FC 1031, 260 F.T.R. 238, 50 Imm. L.R. (3d) 123, 138 A.C.W.S. (3d) 530; Charkaoui v. Canada (Citizenship and Immigration), [2007] 1 S.C.R. 350, [2007] S.C.J. No. 9, 2007 SCC 9, 276 D.L.R. (4th) 594, 358 N.R. 1, J.E. 2007-455, 54 Admin. L.R. (4th) 1, 44 C.R. (6th) 1, 152 C.R.R. (2d) 17, 59 Imm. L.R. (3d) 1, EYB 2007-114995, 154 A.C.W.S. (3d) 363; Chiau v. Canada (Minister of Citizenship and Immigration), 2000 16793 (FCA), [2000] F.C.J. No. 2043, [2001] 2 F.C. 297, 195 D.L.R. (4th) 422, 265 N.R. 121, 102 A.C.W.S. (3d) 178 (C.A.) [Leave to appeal to S.C.C. refused [2001] S.C.C.A. No. 71]; France v. Diab, [2010] O.J. No. 5317, 2010 ONSC 401 (S.C.J.); France v. Diab, [2011] O.J. No. 2398 (S.C.J.); France v. Ouzghar (2009), 94 O.R. (3d) 601, [2009] O.J. No. 277, 2009 ONCA 69, 248 O.A.C. 280, 184 C.R.R. (2d) 92, 252 C.C.C. (3d) 158; Harkat (Re), [2005] F.C.J. No. 481, 2005 FC 393, 261 F.T.R. 52, 45 Imm. L.R. (3d) 65, 138 A.C.W.S. (3d) 346; Ismail (Re), [1998] UKHL 32, [1999] 1 A.C. 320, [1998] 3 W.L.R. 495 (H.L.); Jaballah (Re), [2010] F.C.J. No. 95, 2010 FC 79, [2011] 2 F.C.R. 145, 364 F.T.R. 7, 204 C.R.R. (2d) 262, 2011EXP-3722; Kindler v. Canada (Minister of Justice), 1991 78 (SCC), [1991] 2 S.C.R. 779, [1991] S.C.J. No. 63, 84 D.L.R. (4th) 438, 129 N.R. 81, J.E. 91-1495, 67 C.C.C. (3d) 1, 8 C.R. (4th) 1, 6 C.R.R. (2d) 193, 14 W.C.B. (2d) 30; Lake v. Canada (Minister of Justice), [2008] 1 S.C.R. 761, [2008] S.C.J. No. 23, 2008 SCC 23, 292 D.L.R. (4th) 193, 373 N.R. 339, J.E. 2008-970, 72 Admin. L.R. (4th) 30, 230 C.C.C. (3d) 449, 56 C.R. (6th) 336; [page178] Mahjoub v. Canada (Minister of Citizenship and Immigration), [2006] F.C.J. No. 1862, 2006 FC 1503, [2007] 4 F.C.R. 247, 304 F.T.R. 290, 59 Imm. L.R. (3d) 257, 154 A.C.W.S. (3d) 177; Mahjoub (Re), [2010] F.C.J. No. 900, 2010 FC 787, 373 F.T.R. 36, 90 Imm. L.R. (3d) 76; McVey (Re), 1992 48 (SCC), [1992] 3 S.C.R. 475, [1992] S.C.J. No. 95, 97 D.L.R. (4th) 193, 144 N.R. 81, [1993] 1 W.W.R. 289, J.E. 92-1762, 16 B.C.A.C. 241, 73 B.C.L.R. (2d) 145, 77 C.C.C. (3d) 1, 17 W.C.B. (2d) 585; National Corn Growers Assn. v. Canada (Import Tribunal), 1990 49 (SCC), [1990] 2 S.C.R. 1324, [1990] S.C.J. No. 110, 74 D.L.R. (4th) 449, 114 N.R. 81, J.E. 90-1573, 45 Admin. L.R. 161, 3 T.C.T. 5303, 23 A.C.W.S. (3d) 688; Othman (Abu Qatada) v. United Kingdom, Application No. 8139/09 (January 17, 2012) (Eur. Ct. H.R.); Pastore v. Aviva Canada Inc. (2012), 112 O.R. (3d) 523, [2012] O.J. No. 4508, 2012 ONCA 642, 38 M.V.R. (6th) 177, 296 O.A.C. 281, [2012] I.L.R. I-5350, 12 C.C.L.I. (5th) 1, 355 D.L.R. (4th) 265, 220 A.C.W.S. (3d) 408; R. v. Arcuri, [2001] 2 S.C.R. 828, [2001] S.C.J. No. 52, 2001 SCC 54, 203 D.L.R. (4th) 20, 274 N.R. 274, J.E. 2001-1731, 150 O.A.C. 126, 157 C.C.C. (3d) 21, 44 C.R. (5th) 213, 50 W.C.B. (2d) 512; R. v. Biniaris, [2000] 1 S.C.R. 381, [2000] S.C.J. No. 16, 2000 SCC 15, 184 D.L.R. (4th) 193, 252 N.R. 204, J.E. 2000-838, 134 B.C.A.C. 161, 143 C.C.C. (3d) 1, 32 C.R. (5th) 1, 45 W.C.B. (2d) 454; R. v. G. (A.), [2000] 1 S.C.R. 439, [2000] S.C.J. No. 18, 2000 SCC 17, 184 D.L.R. (4th) 238, 252 N.R. 272, 132 O.A.C. 1, 143 C.C.C. (3d) 46, 32 C.R. (5th) 45, 45 W.C.B. (2d) 455; R. v. Hape, [2007] 2 S.C.R. 292, [2007] S.C.J. No. 26, 2007 SCC 26, 280 D.L.R. (4th) 385, 363 N.R. 1, J.E. 2007-1140, 227 O.A.C. 191, 220 C.C.C. (3d) 161, 47 C.R. (6th) 96, 160 C.R.R. (2d) 1, 73 W.C.B. (2d) 528, EYB 2007-120452; R. v. Yebes, 1987 17 (SCC), [1987] 2 S.C.R. 168, [1987] S.C.J. No. 51, 43 D.L.R. (4th) 424, 78 N.R. 351, [1987] 6 W.W.R. 97, J.E. 87-995, 17 B.C.L.R. (2d) 1, 36 C.C.C. (3d) 417, 59 C.R. (3d) 108; Rizzo & Rizzo Shoes Ltd. (Re) (1998), 1998 837 (SCC), 36 O.R. (3d) 418, [1998] 1 S.C.R. 27, [1998] S.C.J. No. 2, 154 D.L.R. (4th) 193, 221 N.R. 241, J.E. 98-201, 106 O.A.C. 1, 50 C.B.R. (3d) 163, 33 C.C.E.L. (2d) 173, 98 CLLC Â210-006, 76 A.C.W.S. (3d) 894; Saadi v. Italy, [GC] Application No. 37201/06 (February 28, 2008), at para. 140 (Eur. Ct. H.R.); Schreiber v. Canada (Attorney General), [2002] 3 S.C.R. 269, [2002] S.C.J. No. 63, 2002 SCC 62, 216 D.L.R. (4th) 513, 292 N.R. 250, J.E. 2002-1705, 164 O.A.C. 354, 167 C.C.C. (3d) 51, 22 C.P.C. (5th) 207, 116 A.C.W.S. (3d) 372, 54 W.C.B. (2d) 416; Soering v. United Kingdom, Application No. 14038/88 (July 7, 1989), at para. 91 (Eur. Ct. H.R.); Suresh v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 3, [2002] S.C.J. No. 3, 2002 SCC 1, 208 D.L.R. (4th) 1, 281 N.R. 1, J.E. 2002-161, 37 Admin. L.R. (3d) 159, 90 C.R.R. (2d) 1, 18 Imm. L.R. (3d) 1, 110 A.C.W.S. (3d) 1104, revg 2000 17101 (FCA), [2000] F.C.J. No. 5, [2000] 2 F.C. 592, 183 D.L.R. (4th) 629, 252 N.R. 1, 18 Admin. L.R. (3d) 159, 5 Imm. L.R. (3d) 1, 94 A.C.W.S. (3d) 731 (C.A.); United States of America v. Anekwu, [2009] 3 S.C.R. 3, [2009] S.C.J. No. 41, 2009 SCC 41, 275 B.C.A.C. 282, 393 N.R. 77, 310 D.L.R. (4th) 1, 197 C.R.R. (2d) 315, J.E. 2009-1755, EYB 2009-164017, 247 C.C.C. (3d) 99, 96 B.C.L.R. (4th) 1, 69 C.R. (6th) 48, [2009] 11 W.W.R. 383; United States of America v. Burns, [2001] 1 S.C.R. 283, [2001] S.C.J. No. 8, 2001 SCC 7, 195 D.L.R. (4th) 1, 265 N.R. 212, [2001] 3 W.W.R. 193, J.E. 2001-458, 148 B.C.A.C. 1, 85 B.C.L.R. (3d) 1, 151 C.C.C. (3d) 97, 39 C.R. (5th) 205, 81 C.R.R. (2d) 1, REJB 2001-22580, 48 W.C.B. (2d) 400; United States of America v. Dynar (1997), 1997 359 (SCC), 33 O.R. (3d) 478, [1997] 2 S.C.R. 462, [1997] S.C.J. No. 64, 147 D.L.R. (4th) 399, 213 N.R. 321, J.E. 97-1400, 101 O.A.C. 321, 115 C.C.C. (3d) 481, 8 C.R. (5th) 79, 44 C.R.R. (2d) 189, 35 W.C.B. (2d) 8; United States of America v. Kwok, [2001] 1 S.C.R. 532, [2001] S.C.J. No. 19, 2001 SCC 18, 197 D.L.R. (4th) 1, 267 N.R. 310, J.E. 2001-782, 145 O.A.C. 36, 152 C.C.C. (3d) 225, 41 C.R. (5th) 44, 81 C.R.R. (2d) 189, REJB 2001-23416, 49 W.C.B. (2d) 154; United States of America v. Michaelov, [2010] O.J. No. 5226, 2010 ONCA 819, 279 O.A.C. 373, 264 C.C.C. (3d) 480 [Leave to appeal to S.C.C. refused [2011] S.C.C.A. No. 37]; United States of America v. Shephard, 1976 8 (SCC), [1977] 2 S.C.R. 1067, [1976] S.C.J. No. 106, 70 D.L.R. (3d) 136, 9 N.R. 215, 30 C.C.C. (2d) 424, 34 C.R.N.S. 207; [page179] United States of America v. Yang (2001), 2001 20937 (ON CA), 56 O.R. (3d) 52, [2001] O.J. No. 3577, 203 D.L.R. (4th) 337, 149 O.A.C. 364, 157 C.C.C. (3d) 22, 45 C.R. (5th) 205, 87 C.R.R. (2d) 300, 51 W.C.B. (2d) 216 (C.A.)

Statutes referred to

Canadian Charter of Rights and Freedoms, s. 7

Code of Criminal Procedure (France), arts. 134, 175, 176

Criminal Code, R.S.C. 1985, c. C-46, s. 269.1(4)

Extradition Act, S.C. 1999, c. 18 [as am.], ss. 3, (1), 29(1), (a), 32, 33(3), (a), 44(1)(a), 57 [as am.], (7) [as am.]

Extradition Act 1989 (UK), 1989, c. 33

Federal Courts Act, R.S.C. 1985, c. F.7, s. 18.1(4) [as am.]

Immigration Act, R.S.C. 1985, c. I.2, s. 53(1)(b)

Immigration and Refugee Protection Act, S.C. 2001, c. 27 [as am.], s. 83(1)(h), (1.1)

Treaties and conventions referred to

Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, December 10, 1984, 1465 U.N.T.S. 85, art. 15

Convention for the Protection of Human Rights and Fundamental Freedoms, November 4, 1950, 213 U.N.T.S. 221

Extradition Treaty between the Government of Canada and the Government of the Republic of France, November 17, 1988, Can. T.S. 1989 No. 38, art. 3

International Covenant on Civil and Political Rights, December 19, 1966, 999 U.N.T.S. 171, art. 14(3)(e)

Authorities referred to

Côté, Pierre-André, Stéphane Beaulac and Mathieu Devinat, The Interpretation of Legislation in Canada, 4th ed. (Toronto: Carswell, 2011)

Driedger, Elmer, Construction of Statutes, 2nd ed. (Toronto: Butterworths, 1983)

Human Rights Watch, "No Questions Asked": Intelligence Cooperation with Countries that Torture (New York: Human Rights Watch, 2010)

Human Rights Watch, Preempting Justice: Counterterrorism Laws and Procedures in France (New York: Human Rights Watch, 2008)

Iacobucci, Frank, Internal Inquiry Into the Actions of Canadian Officials in Relation to Abdullah Almalki, Ahmad Abou-Elmaati and Muayyed Nureddin (Ottawa: Minister of Public Works and Government Services, 2010)

International Commission of Jurists, Assessing Damage, Urging Action: Report of Eminent Jurists Panel on Terrorism, Counter-terrorism and Human Rights (Geneva: International Commission of Jurists, 2009)

Méndez, Juan E., Report submitted by the Special Rapporteur on Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, Human Rights Council, 16th Sess., U.N. Doc. A/HRC/16/52 (2011)

Nowak, Manfred, Report of the Special Rapporteur on Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, U.N. Doc. A/61/259 (2006)

O'Connor, Dennis R., Report of the Events Relating to Maher Arar (Ottawa: Minister of Public Works and Government Services, 2006)

Ontario Ministry of the Attorney General, Inquiry into Pediatric Forensic Pathology in Ontario (Toronto: Queen's Printer for Ontario, 2008)

United Nations Committee Against Torture, Summary Record of the First Part (Public) of the 424th Meeting, 24th Sess., U.N. Doc. CAT/C/SR.424 (2001)

United Nations Human Rights Committee, General Comment No. 31: Nature of the General Legal Obligation Imposed on States Parties to the Covenant, 80th Sess., U.N. Doc. CCPR/C/21/ Rev.1/Add. 13 (2004) [page180]

APPEAL from the committal order of Maranger J., [2011] O.J. No. 2551, 2011 ONSC 337, 236 C.R.R. (2d) 248 (S.C.J.); APPLICATION for judicial review of the surrender order of the Minister of Justice dated April 4, 2012.

Marlys Edwardh, Daniel Sheppard and Donald Bayne, for applicant/appellant.

Janet Henchey and Jeffrey Johnston, for respondents.

Lorne Waldman, for intervenor Amnesty International.

Brendan van Niejenhuis and Justin Safayeni, for intervenor British Columbia Civil Liberties Association.

Anil Kapoor and Lindsay Daviau, for intervenor Canadian Civil Liberties Association.

BY THE COURT: --

I. Overview

[1] The appellant, Hassan Diab, is wanted in France for his alleged role in a 1980 bombing outside a Paris synagogue that killed four people and injured 40 others. Following a long and sometimes acrimonious hearing, in June 2011 the extradition judge committed the appellant to await surrender. In April 2012, the Minister of Justice ordered the appellant's surrender to French authorities. The appellant appeals from his committal and seeks judicial review of the Minister's decision to surrender.

[2] The case against the appellant is circumstantial. In his reasons on the committal order, the extradition judge held that four of the five pieces of evidence proffered by the Attorney General on behalf of France were insufficient alone to justify committing the appellant for extradition. The remaining piece of evidence, a handwriting analysis that linked the appellant to the suspected bomber, was dismissed by the appellant's experts as methodologically flawed and unreliable. While the extradition judge agreed that the French expert's conclusions were "suspect", he concluded that her evidence should not be completely rejected as "manifestly unreliable" and found it tipped the scale in favour of committal.

[3] On appeal, the appellant submits that the extradition judge took too narrow a view of his role in assessing the reliability of the requesting state's evidence. The appellant submits that, in doing so, the extradition judge misinterpreted the leading case on the test for committal, United States of America v. Ferras; United States of America v. Latty, [2006] 2 S.C.R. 77, [2006] S.C.J. No. 33, 2006 SCC 33, [page181] and the decisions from this court that followed.

[4] Ferras instructs that, while the evidence presented by a requesting state at an extradition hearing is presumptively reliable, s. 7 of the Canadian Charter of Rights and Freedoms requires that the person sought be given a meaningful opportunity to rebut that presumption. The extradition judge cannot act as a rubber stamp: Ferras, at para. 25. This may require the court to engage in a limited weighing of the evidence presented by a requesting state to decide whether committal is justified: Ferras, at para. 46.

[5] In light of this, the appellant argues, the extradition judge erred in holding that the weaknesses in the French handwriting report were matters of ultimate reliability for the French judge to assess. The appellant also seeks leave to introduce fresh evidence to establish that, contrary to the extradition judge's suggestion, there is no unique French methodology for handwriting analysis that could explain what he says is the French expert's markedly divergent approach.

[6] The Canadian Civil Liberties Association ("CCLA") intervenes in support of the appellant's position. It submits that the extradition judge's interpretation of Ferras yields not only an incorrect, but also an unconstitutional, result.

[7] Before the Minister, the appellant resisted surrender on grounds that implicitly impugned France's inquisitorial system of justice. He argued that the Minister lacked jurisdiction to surrender him to France, because France had yet to make a decision that he would stand trial. The Minister rejected his submissions as inconsistent with the principle of comity.

[8] The appellant also argued that it would be unjust, oppressive and contrary to the Charter for the Minister to extradite him on the basis of unsourced intelligence and/or evidence that was plausibly connected to the use of torture. The Minister rejected the appellant's proposed "plausible connection" standard but, in any event, concluded that the appellant had failed to establish such a connection. He also noted that, as a party to the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, December 10, 1984, 1465 U.N.T.S. 85 ("Convention Against Torture"), France has a duty to ensure that evidence derived from torture is not used in any legal proceedings. If the appellant has concerns about the provenance of the intelligence used to build the case against him, the Minister concluded, he may raise them before the French courts, the European Court of Human Rights ("ECHR") and the United Nations Committee Against Torture. [page182]

[9] On the judicial review application, the appellant submits that the Minister misunderstood his submissions, considered irrelevant factors and misapplied the law to reach an unreasonable conclusion.

[10] Without taking a position on the outcome, the intervenor Amnesty International submits that the Minister must refuse extradition where there is a "real risk" that evidence obtained by torture would be admitted at trial. The intervenor the British Columbia Civil Liberties Association ("BCCLA") submits that if the person sought can make out a "plausible connection" between the evidence against him and the use of torture, the Minister must either rebut that connection based on specific information or else satisfy himself or herself that the evidence will not be used against the person sought.

[11] For the reasons that follow, we dismiss the appeal and the application for judicial review.

[12] In our view, the extradition judge did not err in his approach in finding that the French expert's handwriting analysis was not manifestly unreliable. We conclude that the appellant has received the meaningful judicial determination that Ferras requires.

[13] The Minister's surrender decision was reasonable. In our view, the Minister had jurisdiction to order the appellant's surrender even though a trial in France is not a certainty. On our reading of the relevant authorities, an extradition request must be grounded in the anticipation that there will be a trial in the requesting state. In other words, a process or prosecution must have been initiated that will, if not discontinued, lead to a trial. A trial of that person, however, need not be inevitable.

[14] We agree with the appellant and the intervenors that a two-step approach is called for when a person facing surrender alleges that the case against him relies on evidence that is derived from torture. First, the person facing surrender must establish a "plausible connection" between the challenged evidence and the use of torture. This is a relatively low threshold. If that threshold is met, then the Minister is called upon to make further inquiries to determine whether there is a "real risk" that torture-derived evidence will be used in the proposed foreign proceeding. If the Minister concludes that there is a "real risk" that torture-derived evidence will be used in the proposed foreign proceeding, surrender should generally be refused.

[15] Here, while the Minister essentially conflated the two steps of the inquiry and did not expressly use the "real risk" language, it is apparent from the assurances he sought and the [page183] conclusions he reached that he was satisfied this standard had been met. That finding is well supported by the record.

II. Factual Background

(1) The Paris bombing

[16] Early in the evening of October 3, 1980, a bomb exploded outside a synagogue at 24 Rue Copernic in Paris, France. The bomb was apparently timed to go off as worshippers celebrating the holiday of Simchat Torah would be streaming out of the synagogue. Fortunately, services ran late and the crowd on the street was smaller than it might otherwise have been. Still, the impact was devastating: four people were killed and dozens more were wounded.

(2) The investigation

[17] The ensuing investigation led police to believe that the bomb had been planted on a parked motorcycle by a man posing under the pseudonym Alexander Panadriyu. Panadriyu crossed into France from Spain on a fake Cypriot passport and checked into the Celtic Hotel near Rue Copernic on September 22, 1980. In what would prove to be the key piece of evidence in the case, Panadriyu signed the hotel registration card. He wrote five words, four in block letters: PANADRIYU, ALEXANDER, LANARCA, technician, CYPRUS. There was also a date written on the registration card. Composite sketches of Panadriyu were made based on physical descriptions given by various people who saw him at the hotel and at the shop where he bought the motorcycle used in the attack.

[18] In the course of the investigation, detectives discovered that Panadriyu had been detained a few days before the bombing for stealing a pair of pliers. Panadriyu signed a police report prior to being released. He was not photographed.

[19] In 1982, a judicial inquiry into the Paris bombing received a report from the French intelligence service stating that the buyer of the motorcycle was a man named Hassan, and that he acted on behalf of a splinter group of the Popular Front for the Liberation of Palestine ("PFLP").

[20] The investigation lay dormant for many years. Then in 1999, the French intelligence service passed on information from an unnamed source which identified several people involved in the bombing. The bomber was identified as Hassan Diab.

[21] After further investigation, in 2008 France requested Mr. Diab's surrender to face charges of murder, attempted murder and mischief. In 2009, Canada issued an authority to proceed. [page184]

(3) The person sought

[22] Hassan Diab was born in Lebanon on November 20, 1953. He arrived in the United States in 1987, and obtained a Ph.D. in sociology from Syracuse University in New York. He became a Canadian citizen in 2006. At the time of his arrest, Mr. Diab held contract positions as a lecturer at Carleton University and the University of Ottawa. He has no prior criminal record. He has the unrelenting support of many of his friends and colleagues from Carleton.

III. Appeal from Committal

A. Committal for extradition: An overview

(1) General principles

[23] As this court observed in United States of America v. Yang (2001), 2001 20937 (ON CA), 56 O.R. (3d) 52, [2001] O.J. No. 3577 (C.A.), at para. 31, "[t]here is no single universal model for extradition in all countries or even in the same country".

[24] On the request of any of more than 80 states or other entities -- ranging from the United States to Haiti, Tonga and Zimbabwe to the International Criminal Court -- a person may be extradited from Canada to be prosecuted or sentenced. Extradition must be in accordance with the Extradition Act, S.C. 1999, c. 18 (the "Act") and the relevant extradition agreement. Under the Act, extradition can only be sought for serious offences -- generally ones punishable in Canada by imprisonment for two or more years.

[25] The provisions of the extradition agreements vary. Some date back to the 19th century and were entered into with colonial powers. Others are more recent and reflect Canada's modern international relations.

[26] The request in this case was made by the Republic of France pursuant to a treaty entered into in 1989: Extradition Treaty between the Government of Canada and the Government of the Republic of France, November 17, 1988, Can. T.S. 1989 No. 38. Article 3 of that treaty provides that Canada and France "shall not be bound to extradite [their] own nationals. Nationality shall be determined as of the date of the offence for which extradition is requested."

[27] The appellant was not a Canadian national in 1980, when the alleged offence for which extradition is sought occurred. Accordingly, subject to the provisions of the Act and the other provisions of the treaty, Canada is bound to extradite him. [page185]

(2) The statutory framework

[28] Section 29(1)(a) of the Act provides:

29(1) A judge shall order the committal of the person into custody to await surrender if

(a) in the case of a person sought for prosecution, there is evidence admissible under this Act of conduct that, had it occurred in Canada, would justify committal for trial in Canada[.]

[29] Evidence at an extradition hearing is proffered by way of a "record of the case" filed by the requesting state. Section 32 of the Act provides that evidence that would not otherwise be admissible under Canadian law is admissible at an extradition hearing if the contents of the documents contained in a record of the case are certified under s. 33(3).

[30] That section provides:

33(3) A record of the case may not be admitted unless

(a) in the case of a person sought for the purpose of prosecution, a judicial or prosecuting authority of the extradition partner certifies that the evidence summarized or contained in the record of the case is available for trial and

(i) is sufficient under the law of the extradition partner to justify prosecution, or

(ii) was gathered according to the law of the extradition partner[.]

[31] In this case, the French juge d'instruction certified that the evidence in the record of the case was available for trial and was gathered according to French law. He did not certify that the evidence summarized or contained in the record of the case was sufficient under French law to justify prosecution.¹

(3) The governing jurisprudence

(i) Ferras

[32] In Ferras, McLachlin C.J.C., writing for a unanimous Supreme Court, held, at para. 46, that the direction in s. 29(1) that the extradition judge must decide if there is evidence that would "justify committal" requires the judge to assess "whether admissible evidence shows the justice or rightness in committing a person for extradition" (emphasis in original). If a judge on an extradition hearing concludes that the evidence is manifestly [page186] unreliable, the judge should not order extradition under s. 29(1): at para. 40. Section 7 of the Charter demands that the extradition judge have the discretion to refuse committal on insufficient evidence: at paras. 41-43.

[33] Ferras reversed the pre-Charter case of United States of America v. Shephard, 1976 8 (SCC), [1977] 2 S.C.R. 1067, [1976] S.C.J. No. 106, where a majority of the Supreme Court held that an extradition judge must commit a person for surrender if there is any evidence on all the elements of the offence, no matter how suspect or dubious.

[34] Chief Justice McLachlin set out the test for committal under s. 29(1), at para. 46:

It is not enough for evidence to merely exist on each element of the crime. The evidence must be demonstrably able to be used by a reasonable, properly instructed jury to reach a verdict of guilty. If the evidence is incapable of demonstrating this sufficiency for committal, then it cannot "justify committal". The evidence need not convince an extradition judge that a person sought is guilty of the alleged crimes. That assessment remains for the trial court in the foreign state. However, it must establish a case that could go to trial in Canada. This may require the extradition judge to engage in limited weighing of the evidence to determine, not ultimate guilt, but sufficiency of evidence for committal to trial.

(Emphasis in original)

[35] The chief justice proceeded to explain that, following from principles of international comity, certification under s. 33(3)(a) raises a presumption that the evidence in a record of the case is reliable. "Certification," she held, "is the indicium of reliability that Parliament has prescribed for evidence in these circumstances. Unless challenged, certification establishes reliability": at para. 52. However, the person sought may challenge the sufficiency of the case, including the threshold reliability of certified evidence.

[36] The chief justice set out the procedure to be followed, at para. 54:

Challenging the justification for committal may involve adducing evidence or making arguments on whether the evidence could be believed by a reasonable jury. Where such evidence is adduced or such arguments are raised, an extradition judge may engage in a limited weighing of evidence to determine whether there is a plausible case. The ultimate assessment of reliability is still left for the trial where guilt and innocence are at issue. However, the extradition judge looks at the whole of the evidence presented at the extradition hearing and determines whether it discloses a case on which a jury could convict. If the evidence is so defective or appears so unreliable that the judge concludes it would be dangerous or unsafe to convict, then the case should not go to a jury and is therefore not sufficient to meet the test for committal.

(Emphasis added) [page187]

(ii) Thomlison and Anderson

[37] United States of America v. Thomlison (2007), 84 O.R. (3d) 161, [2007] O.J. No. 246, 2007 ONCA 42, leave to appeal refused [2007] S.C.C.A. No. 179, and United States of America v. Anderson (2007), 85 O.R. (3d) 380, [2007] O.J. No. 449, 2007 ONCA 84, leave to appeal to S.C.C. refused [2007] S.C.C.A. No. 159, released within weeks of each other, were the first decisions from this court to interpret and apply the Supreme Court's decision in Ferras. In both cases, the court observed that Ferras exposed a tension between the purpose of extradition proceedings and the demands of the Charter. On the one hand, extradition hearings are meant to be expeditious and to facilitate prompt compliance with Canada's treaty obligations: Anderson, at para. 42; United States of America v. Dynar (1997), 1997 359 (SCC), 33 O.R. (3d) 478, [1997] 2 S.C.R. 462, [1997] S.C.J. No. 64, at para. 122. On the other hand, s. 7 of the Charter requires that the person sought must be able to challenge the case against him through a meaningful judicial process: Ferras, at para. 26.

[38] Accepting that Ferras empowered extradition judges to assess the quality of the evidence put forth by the requesting state to justify committal, the question became, as Moldaver J.A. put it in Thomlison, at para. 6: "[T]o what extent can extradition judges now weigh the evidence in deciding whether the test for committal has been met?" (emphasis in original).

[39] He answered, at para. 45:

Unlike the situation that existed post Shephard, Ferras now authorizes extradition judges to assess the availability and quality of the evidence that can legitimately be included in the "some evidence" basket for sufficiency purposes. In my view, that enables them to discard evidence that is not realistically available for trial and/or evidence that is manifestly unreliable, i.e. evidence upon which it would clearly be dangerous or unsafe to convict. It does not allow extradition judges to refuse to commit where there is "available and reliable" evidence in the "some evidence" basket upon which a reasonable jury, properly instructed, could convict.

(Emphasis added)

[40] In Anderson, the record of the case included a statement by a co-conspirator seeking a favourable bargain with the prosecuting authorities. Counsel for the person sought did not argue that the obvious frailties in the co-conspirator's evidence would justify giving the statement no weight. Rather, he argued he should have been able to call the co-conspirator at the extradition hearing. He submitted that had he done so, he may have succeeded in impeaching the co-conspirator's credibility or the reliability of his statement. Justice Doherty concluded that [page188] the extradition judge correctly held that the co-conspirator could not be called as a witness.

[41] He elaborated, at para. 28:

. . . Ferras, supra, does not envision weighing competing inferences that may arise from the evidence . . . [or deciding] whether a witness is credible or his or her evidence is reliable . . . . There is no power to deny extradition in cases that appear to the extradition judge to be weak or unlikely to succeed at trial.

[42] Justice Doherty described the scenarios in which an extradition judge might reject evidence as insufficient to justify committal, at para. 30:

Evidence may be rendered "so defective" or "so unreliable" as to warrant disregarding it due to problems inherent in the evidence itself, problems that undermine the credibility or reliability of the source of the evidence, or a combination of these two factors. I would stress, however, that it is only where the concerns with respect to the reliability of the evidence, whatever the source or sources, are sufficiently powerful to justify the complete rejection of the evidence, that these concerns become germane to the s. 29(1)(a) inquiry.

[43] Justice Doherty's explanation that evidence is manifestly unreliable where the concerns as to its reliability justify its complete rejection is not a new one. It is rooted in the dissenting reasons of Spence J. in Shephard, which were essentially adopted by the Supreme Court in Ferras. At p. 1076 S.C.R. of Shephard, Spence J. endorsed the extradition judge's description of what would constitute "manifestly unreliable" evidence:

A finding that evidence is "manifestly unreliable" or "dubious" does, of course, necessarily involve some sort of weighing process, not, however, for the purpose of determining whether such evidence "proves" the charge but rather for the purpose of determining whether it has any weight at all which could prove the charge.

(Emphasis added)

B. The committal hearing

(1) The evidence against the appellant

[44] The record of the case presented to the extradition judge relied on five pieces of evidence² to connect the appellant to the Paris bombing: (1) the appellant's Lebanese passport dated 1980, with stamps indicating he entered Spain shortly before the [page189] bombing and left shortly after; (2) evidence of his membership in the PFLP; (3) eyewitness descriptions of Panadriyu; (4) composite sketches of Panadriyu that bore some resemblance to the appellant; and (5) an expert analysis report by Mme Anne Bisotti that compared the appellant's handwriting to Panadriyu's writing and concluded that there was a "strong presumption" that the appellant was the author of the hotel registration card (the "Bisotti report").

[45] The extradition judge held that the first four pieces of evidence, whether taken individually or viewed as a whole, were insufficient to justify committing the appellant for extradition. "At best", the extradition judge explained, "they create a certain degree of suspicion concerning his involvement in the terrorist bombing": at para. 188. That left the Bisotti report.

(2) The Bisotti report

[46] The Bisotti report was one of three handwriting analyses contained in the record of the case. The first two were prepared by "handwriting comparison experts" Mme Evelyne Marganne and Mme Dominique Barbe-Prot. These reports compared the writing on the hotel registration card signed by the suspected bomber Panadriyu to samples of what were supposedly the appellant's writing from his days as a graduate student at Syracuse University. Both reports concluded that the appellant was likely the author of the hotel registration card.

[47] In late 2009, several months before the start of the committal hearing, the appellant brought an application to introduce evidence to establish that the Marganne and Barbe-Prot reports were manifestly unreliable. The appellant sought to call evidence from four people whom the extradition judge described as "arguably" among "the world's leading experts in the field of handwriting analysis": Brian Lindblom, John Paul Osborn, Robert Radley and Dan Purdy.

[48] All of these experts were extremely critical of the French experts' methodology. Most importantly, they all concluded that the Marganne and Barbe-Prot reports were completely unreliable because some of the handwriting samples used for comparison were not, in fact, written by the appellant. They were written by his then-wife, Nawal Copty. The extradition judge allowed all four expert reports to be filed and permitted two of the four experts to be called.

[49] Five months later, France withdrew the Marganne and Barbe-Prot reports from the record of the case and replaced them with the Bisotti report. Again, the appellant sought to introduce evidence to challenge the reliability of the handwriting [page190] analysis. The extradition judge [ [2010] O.J. No. 5317, 2010 ONSC 401 (S.C.J.)] acknowledged the "danger" that doing so would "transform the extradition proceeding into a criminal trial by impermissibly weighing competing inferences derived from expert evidence". He explained [at para. 14]: "The Extradition Act legislates that the Bisotti report is admissible and presumptively reliable; I am not allowed to weigh its ultimate reliability." On the other hand, he held [at para. 17] that "the line here between expert opinion evidence being used to support a competing inference or as evidence demonstrating manifest unreliability is blurred". The extradition judge allowed the appellant to introduce technical reviews of the Bisotti report from Mr. Lindblom, Mr. Osborn and Mr. Radley.

[50] As the extradition judge recounted, at para. 91 of the committal decision, Mme Bisotti is the head of the documents, handwriting and traces sections of the laboratory of the police scientifique of Paris. She has been an expert in handwriting analysis since 1993, and has been consulted on hundreds of cases. Her stated mandate was to opine on whether the appellant "is certainly or may be (in which case, try to indicate a degree of probability) the writer of the five words and date contained on the hotel registration card, and the signature at the bottom of the police report": at para. 91.

[51] Mme Bisotti compared the hotel registration card and the signature on the police report to exemplars of the appellant's handwriting from Syracuse University and pages taken from the appellant's American immigration file. Those exemplars dated from between 1994 and 1998. Applying what she described as the ENFHEX (European Network of Forensic Handwriting Experts) methodology, she concluded that there was a "strong presumption" that the appellant was the author of the five words on the hotel registration card, a "weak presumption" that he was the author of the date on the hotel registration card and a "presumption" that he was the author of the signature on the police report.

(3) The critiques of the Bisotti report

[52] The appellant's experts attacked the Bisotti report as completely methodologically unsound. As one example, they said Mme Bisotti purported to establish authorship of the hotel registration card in part by counting the similarities between the two sets of handwriting. They all maintained that in handwriting identification it is the differences that establish identification. Equally problematic, they said, Mme Bisotti dismissed the [page191] differences between the two sets of handwriting as "natural variations" without any evidentiary basis for doing so.

[53] The appellant's experts were strongly of the view that neither the five words on the Celtic Hotel registration card nor the exemplars of the appellant's handwriting provided a sufficient basis for proper analysis. The hotel registration card, they said, provided such a small sample of the bomber's handwriting as to be basically worthless. This was especially so because Panadriyu printed in block letters, presumably in an effort to disguise his natural handwriting. Meanwhile, the university papers and the pages from the appellant's immigration file -- dating from the 1990s -- were of limited use because they were written so long after the subject documents.

[54] The appellant's experts were also fiercely critical of Mme Bisotti's "mandate", which they interpreted to preclude the finding that Hassan Diab was not the author of the hotel registration card. In addition to arguing that the report was inherently biased, the appellant's experts also challenged Mme Bisotti's qualifications. As we discuss below, the appellant abandoned those arguments on appeal.

[55] Each of the appellant's experts summarized their opinions on the merits of the Bisotti report in extremely harsh terms. In short, they all opined that no competent handwriting analyst would have reached the same conclusions as Mme Bisotti.

(4) The parties' submissions on whether the Bisotti report was "manifestly unreliable"

[56] In light of the concerns raised about the Bisotti report, the appellant argued that it could not meet the standard for threshold admissibility as described in R. v. Mohan, 1994 80 (SCC), [1994] 2 S.C.R. 9, [1994] S.C.J. No. 36 and R. v. Abbey (2009), 97 O.R. (3d) 330, [2009] O.J. No. 3534, 2009 ONCA 624, leave to appeal to S.C.C. refused [2010] S.C.C.A. No. 125, and addressed in Justice Goudge's report of the Inquiry into Pediatric Forensic Pathology in Ontario (Toronto: Queen's Printer for Ontario, 2008) ("Goudge report"). Those principles include:

the trial judge is the gatekeeper who must vigilantly control the admission of expert evidence;

the expert must be adequately qualified and scrupulously impartial; and

threshold reliability includes careful consideration of the methodology used by a given expert. [page192]

[57] The extradition judge described the appellant's argument this way, at para. 96: "[I]f Ms. Bisotti's report would be found to lack threshold reliability in a Canadian court, then it should be considered manifestly unreliable evidence for the purposes of an extradition proceeding."

[58] In response, the Attorney General on behalf of France submitted that the appellant's evidence was simply evidence of competing inferences that went to ultimate, not threshold, reliability. The Attorney General noted that, by virtue of its certification under s. 33(3) of the Act, the Bisotti report was already presumptively reliable. The Attorney General submitted that threshold reliability was therefore established; "manifest unreliability" means evidence that is, or can be shown to be, totally unreliable.

(5) The extradition judge's decision that the Bisotti report is not manifestly unreliable

[59] The extradition judge first ruled orally on this issue on February 18, 2011, before the conclusion of the extradition hearing. At that time, he provided an outline of his reasoning supporting his decision to reject the appellant's application to exclude the Bisotti report from the record of the case and explained that his full reasons would be contained in his final written decision. In his oral reasons, the extradition judge described the issue before him as whether the Bisotti report "should be excluded from the Record of the Case as manifestly unreliable, or as so defective that it would be dangerous to convict upon [it]".

[60] At the outset of his consideration of this issue in his written reasons, at para. 80, the extradition judge cited the governing principle, taken from this court's decision in United States of America v. Michaelov, [2010] O.J. No. 5226, 2010 ONCA 819, 264 C.C.C. (3d) 480, at para. 47, leave to appeal to S.C.C. refused [2011] S.C.C.A. No. 37: the only means available to rebut the presumption of threshold reliability is to demonstrate that the evidence is "manifestly unreliable" or "so defective" that it should be removed from consideration in deciding the issue of committal. He rejected the appellant's submission that the Abbey framework could be used to rebut the presumption of threshold reliability. He explained, at para. 101, that to accept the appellant's submission "would necessitate the imposition of Canadian standards of admissibility of expert opinion evidence on the requesting state". The extradition judge held that doing so would offend the Act. That statute has its own specific rules of evidence that permit the admission of evidence even if it would not be admitted under Canadian law. [page193]

[61] As to the substance of the appellant's arguments, the extradition judge wrote that the report "has been shown to be based on some questionable methods and on an analysis that seems very problematic" and "is susceptible to a great deal of criticism and attack": at paras. 118, 120. He described the report as "convoluted, very confusing, with conclusions that are suspect": at para. 121.

[62] He found that Mme Bisotti's comparison of the appellant's signature with the "illegible fictitious signature" on the police report seemed illogical. And he observed what he considered an important difference between the samples of the appellant's handwriting relied on by Mme Bisotti and the letters on the hotel registration card that she did not discuss in her report. Significantly, on the other hand, the extradition judge agreed that what Mme Bisotti described as an important identifier -- "the slopes and slants within words" -- clearly existed in the samples and the hotel registration card.

[63] The extradition judge concluded that the Bisotti report was not "manifestly unreliable". Accepting the Attorney General's submissions, he explained that information contained in a record of the case is presumptively reliable for purposes of committal. While the appellant's experts had exposed major weaknesses in the Bisotti report, the extradition judge concluded that choosing between the experts' competing views was a task for the trier of fact and a question of ultimate, not threshold, reliability: at paras. 122-23.

[64] In coming to this conclusion, the extradition judge relied on this court's decisions in Thomlison, Anderson and Michaelov.

[65] The extradition judge held that the parties' opposing views of Mme Bisotti's objectivity raised competing inferences. While he saw merit in the appellant's experts' criticisms, the extradition judge determined that to accept those criticisms would require him to weigh competing inferences and perform the work of the trier of fact. "It is the stuff of a trial", he concluded, at para. 106.

[66] With regard to methodology, the extradition judge wondered (as he had in his earlier oral ruling) whether the markedly divergent opinions between the French experts and the appellant's experts (from Canada, the U.S. and the U.K.) were indicative of a unique French methodology. In remarks that form the basis of the appellant's fresh evidence application, he explained, at para. 112:

While I found that the three experts' evidence on appropriate methodology was convincing in that it appeared to be the logical approach to comparing handwriting, I cannot necessarily conclude that their criticisms respecting methodology related exclusively to the report of Ms. Bisotti. There were [page194] a total of three reports submitted in the area of handwriting comparison by three different French experts: Ms. Barbe-Prot, Ms. Marganne and Ms. Bisotti. The Diab experts criticized the methodology employed by all three. It seems to me that this leaves room for the possibility that the Republic of France does have a different approach/methodology in relationship to handwriting comparison analysis.

[67] And, at para. 116, he explained what "manifest unreliability" means in the context of an extradition case:

It seems to me that manifest unreliability in the context of an extradition case would mean evidence that is so devoid of reliability that it should be rejected out of hand when considering the issue of committal. It is evidence that is so unreliable that it is not worthy of consideration by the trier of fact. A high threshold attaches to the categorization of evidence that may be called "manifestly unreliable."

[68] The extradition judge noted that handwriting comparison is a soft science that Canadian courts have long recognized as an admissible form of expert evidence. He continued, at paras. 122 and 123:

The bottom line is that very strong criticism, coupled with competing inferences from other experts, does not render another expert's opinion manifestly unreliable in the context of an extradition.

The complaints levied against the report and its findings, when all is said and done, go the ultimate reliability of the opinion, which is a matter for trial.

[69] The extradition judge explained what, in his view, the appellant would have had to have shown to establish that the Bisotti report was manifestly unreliable, at para. 124:

If the only available conclusion derived from the [record of the case] and the evidence from the three experts presented on behalf of the person sought was that the French expert was biased, unqualified, and used improper methodology in every respect, then a finding of manifest unreliability would be possible. However, the evidence heard at this hearing did not support such an unequivocal finding.

[70] The extradition judge then gave an example of the type of evidence that could have led to a finding of manifest unreliability, at para. 125:

[P]erhaps the clearest way to describe manifestly unreliable evidence is to provide an example. In truth at one point in time it may have existed in this case. This would be in relation to the first two handwriting reports [i.e. the Marganne and Barbe-Prot reports]. If I had found as a fact that these two experts had used the wrong known handwriting to arrive at their conclusions respecting the authorship of the hotel card i.e. (Nawal Copty's instead of Hassan Diab's) that would have amounted to, as the Court in Michaelov, supra, put it "problems inherent in the evidence, problems that undermine the credibility of the source of the evidence or a combination of both factors" that render the evidence manifestly unreliable. [page195]

(6) The extradition judge's decision on the test for committal

[71] Having concluded that the Bisotti report could not be dismissed as manifestly unreliable, the extradition judge proceeded to determine whether there was sufficient evidence to justify the appellant's committal.

[72] The parties differed on whether, if the extradition judge found the Bisotti report was not manifestly unreliable, Ferras permitted the extradition judge to nonetheless weigh the Bisotti report and the other evidence in the record of the case in determining whether or not to commit the appellant for extradition.

[73] The appellant submitted that after completing his manifest unreliability inquiry, the extradition judge should proceed in the same way as an appellate court reviewing an allegation of an unreasonable verdict. In other words, the extradition judge should then ask whether a properly instructed jury, acting judicially, could reasonably convict on the evidence: see R. v. Yebes, 1987 17 (SCC), [1987] 2 S.C.R. 168, [1987] S.C.J. No. 51. Or, as the extradition judge expressed the appellant's position, could the evidence or the lack of evidence lead to an unreasonable verdict?

[74] The extradition judge rejected the appellant's approach. In his view, this approach would necessarily require him to "become engaged in the weighing of evidence" when deciding whether or not to commit the person: at para. 133. Guided by this court's decisions in Thomlison, Anderson and Michaelov, the extradition judge stated the test this way, at para. 139: "Is there evidence in the [record of the case] that is available for trial and not manifestly unreliable, upon which a reasonable jury, properly instructed, could convict?"

[75] The extradition judge framed this as a two-stage process: first, the court determines whether the evidence in the record of the case is unavailable for trial or manifestly unreliable. This requires a limited weighting of the evidence. Second, the court considers the remainder of the evidence to determine whether to commit the person sought. At this second stage, the extradition judge held, the court's ability to weigh the evidence is limited to assessing whether the circumstantial evidence, if any, is reasonably capable of supporting the inferences that the requesting state seeks to have drawn. The court does not go further to consider whether it would conclude that the accused is guilty or to draw the factual inferences or assess credibility: at paras. 140-41.

[76] The extradition judge held, at para. 142, that, on his reading, [page196]

[O]ur Court of Appeal does not contemplate the extradition judge analyzing or weighing the evidence that remains in the [record of the case] to determine whether or not it would be dangerous to convict upon that evidence. The only question to ask at this stage is whether a properly instructed jury acting reasonably could convict on that evidence.

[77] Applying this test to the evidence in the record of the case, the extradition judge concluded that, whether taken individually or viewed as a whole, the passport, the evidence of the appellant's membership in the PFLP, and the eyewitness descriptions and composite sketches would not be sufficient to justify committing the appellant. "At best", he held, "they create a certain degree of suspicion" concerning the appellant's involvement in the bombing: at para. 188.

[78] "The evidence that tips the scale in favour of committal", the extradition judge continued, "is the handwriting comparison evidence. Once found to be reliable for purposes of extradition, i.e., not manifestly unreliable evidence, the question became whether a jury considering the handwriting evidence together with the other evidence in the [record of the case], could find as a fact that Mr. Diab was Alexander Panadriyu and thus one of the persons responsible for the bombing. The short answer is yes. Consequently, when all is said and done, a committal order is warranted": at para. 189.

[79] While the extradition judge was of the view that the case presented by the Republic of France was weak, and the prospects of conviction in the context of a fair trial unlikely, he concluded, at para. 195: "They have presented a prima facie case against him which justifies his having to face a trial in that country."

C. Motion to adduce fresh evidence

(1) Background

[80] More than two months after the extradition hearing was completed, and 12 days before the extradition judge was scheduled to render his committal decision, the appellant brought an application to adjourn the scheduled date for release of the extradition judge's decision in order to perfect an application to re-open the hearing and call new evidence. The new evidence included recently sourced but not yet available expert evidence dealing with the methodology for handwriting comparison used in France. As discussed above, in an oral ruling dismissing the appellant's application to have the Bisotti report excluded because it was manifestly unreliable, the extradition judge had pointed to the fact that "none of the experts could comment on whether or not there is or isn't a different type of methodology [page197] used in the Republic of France". The appellant argued that the new evidence would impact on the extradition judge's ruling that the Bisotti report was not manifestly unreliable.

[81] The extradition judge denied the application. He concluded that the proposed new expert evidence would not change his ruling that the appellant's expert evidence went to the ultimate reliability of the Bisotti report.

[82] The appellant nonetheless proceeded to obtain the evidence of two further handwriting experts, Prof. Pierre Margot and Dr. Raymond Marquis. Although Swiss nationals, Prof. Margot and Dr. Marquis are familiar with forensic document examination in France. It is their evidence that the appellant now seeks leave to adduce as "fresh evidence".

[83] Their report makes clear that there is no different French method of handwriting analysis. Rather, the methodology for handwriting comparison is universal, ENFHEX recommendations correspond to this universal approach and French handwriting analysis is conducted in accordance with the ENFHEX guidelines. Like the appellant's other experts, Prof. Margot and Dr. Marquis say that Mme Bisotti failed to adhere to this methodology.

[84] However, their report also acknowledges that Mme Bisotti is recognized as one of the two major handwriting experts in France. And while the report takes issue with the "leading" nature of the questions that Mme Bisotti was required to answer, Dr. Marquis testified that he did not view Mme Bisotti as biased. Perhaps because of this, on appeal the appellant does not challenge the extradition judge's conclusion that the issues of inherent bias and Mme Bisotti's qualifications were matters for trial, and did not render the Bisotti report manifestly unreliable.

[85] The appellant filed the Margot and Marquis report with the Minister as part of his submissions contesting surrender. Their report therefore forms part of the record before this court on the combined appeal and application for judicial review.

(2) The parties' submissions

[86] The appellant does not argue that the extradition judge erred in refusing his request for a further adjournment to permit him to obtain further evidence. Rather, he brings this motion to admit the evidence subsequently obtained as fresh evidence. He submits that this proposed fresh evidence meets the four criteria in R. v. Palmer, 1979 8 (SCC), [1980] 1 S.C.R. 759, [1979] S.C.J. No. 126 for the admission of fresh evidence.

[87] The only disputed criteria are whether the proposed evidence is fresh, and whether it is relevant. The appellant argues that the evidence could not have been adduced at the committal [page198] hearing because the notion of a unique French system was only raised by the Attorney General's cross-examination of the appellant's experts. The appellant also argues that the fresh evidence could reasonably have been expected to affect the result. He asserts that the extradition judge's comment that evidence of this sort would not have changed his decision with respect to manifest unreliability was driven by the extradition judge's mistaken imposition of too high a standard for "manifest unreliability".

[88] The Attorney General counters that the proposed fresh evidence does not meet the test for admitting fresh evidence on a committal appeal in extradition proceedings. It is not "fresh"; the appellant unsuccessfully applied to adjourn the date set by the extradition judge to render his decision to seek to tender this very evidence. And it is not relevant; like the appellant's other expert evidence, it goes to the ultimate reliability of the Bisotti report.

(3) Analysis

[89] We agree with the Attorney General that the proposed fresh evidence does not meet the test for admission. The extradition judge considered what the appellant advised him was the substance of the proposed evidence and determined that such evidence would not affect his decision.

[90] As the extradition judge said in his reasons dismissing the appellant's application for an adjournment to gather Prof. Margot and Dr. Marquis' evidence, "on the issue of reliability, I find it difficult to conceive that the French experts will be capable of being more critical than the experts already presented by the person sought over the course of approximately 30 days of court time".

[91] It is also worth noting that the extradition judge's decision dismissing the application for an adjournment was delivered just 11 days before he released his 195-paragraph committal decision. It is reasonable to presume that the extradition judge's analysis was substantially advanced by that time and so he was in a position to know with certainty that the proposed new evidence would not alter his view that the reliability of the Bisotti report was a matter for trial.

[92] The substance of the evidence of Prof. Margot and Dr. Marquis is as the appellant advised the extradition judge. Concluding, as we do below, that the extradition judge correctly understood and applied the standard for assessing manifest unreliability, such evidence could not reasonably have affected the result. It follows that the appellant's motion to introduce the [page199] Margot and Marquis report as fresh evidence on the committal appeal is dismissed.

D. Issues on the committal appeal

[93] The appellant frames the issue on the committal appeal as this: did the extradition judge err by ordering committal on the basis of the methodologically unsound Bisotti report, contrary to Ferras?

[94] His submissions can be broken down into four discrete alleged errors.

[95] First, he argues that an expert opinion that does not meet the threshold reliability test in Mohan and Abbey is manifestly unreliable and should not be considered by a judge at an extradition hearing. He submits that the extradition judge erred in law in concluding to the contrary. Applied to this case, the appellant argues that the Bisotti report does not meet the threshold reliability standard of Abbey and is therefore manifestly unreliable.

[96] Second, he argues that the extradition judge's conclusion that the Bisotti report was not manifestly unreliable was rooted in speculation and a misapprehension of the evidence. The extradition judge discounted evidence from all of the appellant's experts that Mme Bisotti used a flawed and totally unreliable methodology on the basis that it was possible that France has a different approach to, or methodology for, handwriting analysis. The evidence before the extradition judge was that there is one accepted method of handwriting analysis used around the world.

[97] Third, the appellant submits that even if the threshold reliability test in Abbey is not applicable in an extradition context, the extradition judge imposed too high a standard for "manifest unreliability". The appellant argues that the extradition judge's approach makes it nearly impossible to establish that an expert report in the record of a case is manifestly unreliable.

[98] Finally, the appellant renews an argument he made to the extradition judge. He submits that, having concluded that the Bisotti report was not manifestly unreliable, the extradition judge failed to ask the second question that Ferras requires: on all of the evidence in the record of the case, would it be dangerous or unsafe to convict the appellant? Relying on United States of America v. Graham, [2007] B.C.J. No. 1390, 2007 BCCA 345, 222 C.C.C. (3d) 1, leave to appeal to S.C.C. refused [2007] S.C.C.A. No. 467, he argues that the extradition judge must apply an "unreasonable verdict/ conviction" test in deciding the issue of committal. The extradition judge acknowledged [page200] that France's case turned on the Bisotti report. The appellant submits that given the problems inherent in the Bisotti report, the answer to such a question would be yes, and the extradition judge should not have committed the appellant.

[99] The CCLA echoes the appellant's final argument. It asserts that committal should not be ordered if the extradition judge concludes, after looking at the whole of the available and reliable evidence, that it would be dangerous or unsafe to convict.

E. Analysis of issues on the committal appeal

[100] We begin by noting that, absent an error in the application of the Ferras standard, a misapprehension of the evidence or an unreasonable conclusion, the extradition judge's determination that the Bisotti report is not manifestly unreliable and that the appellant should be committed involves the exercise of judicial discretion and is entitled to deference: Michaelov, at para. 72.

[101] With that observation, we turn to address each of the extradition judge's alleged errors.

(1) Threshold reliability in Mohan and Abbey versus manifest unreliability in the extradition context

[102] While our reasoning is different, we agree with the extradition judge that the threshold reliability inquiry required by Mohan and Abbey is different than the inquiry that an extradition judge must undertake when the person sought challenges an expert opinion that forms part of the record of the case.

[103] In Mohan, decided more than a decade before Ferras, the Supreme Court highlighted the danger that a jury is apt to accept an expert opinion "as being virtually infallible and as having more weight than it deserves": at p. 21 S.C.R. In Abbey, decided some three years after Ferras, this court provided further guidance about the "cost-benefit" or "gatekeeper" analysis that must be undertaken to determine if an expert opinion should be admitted.

[104] On the benefit side, the potential probative value of the expert opinion must be considered. That assessment requires a consideration of the reliability of the opinion. And, as Doherty J.A. wrote in Abbey, at para. 87: "Reliability concerns reach not only the subject matter of the evidence, but also the methodology used by the proposed expert in arriving at his or her opinion, the expert's expertise and the extent to which the expert is shown to be impartial and objective." On the cost side, the greatest risk is that the jury will be unable to critically assess the expert evidence and will abdicate its fact-finding role to the expert. [page201]

[105] Keeping in mind the dangers posed by expert opinion evidence, the court must determine whether the opinion is sufficiently reliable to be considered by the jury. The trial judge should not be concerned with the ultimate reliability of the evidence. A flexible approach to the determination of reliability at the gatekeeper stage is essential. The factors relevant in determining the reliability of the opinion will differ depending on the nature of the opinion; there is no bright line rule. The gatekeeper inquiry requires an exercise of judicial discretion. As Doherty J.A. wrote, at para. 79: "Different trial judges, properly applying the relevant principles in the exercise of their discretion, could in some situations come to different conclusions on admissibility."

[106] The lens through which a challenged expert opinion in the record of the case must be evaluated is different. Abbey seeks to protect the trier of fact from material that is ultimately detrimental to the truth-seeking function because it might overwhelm the fact-finder or cannot be effectively assessed. Its purpose is to protect the fairness of the trial process. That is not the focus of an extradition judge reviewing the record of the case. Protecting trial fairness is generally the responsibility of the requesting state, not the extradition judge. The extradition judge reviews the record of the case to ensure the minimum constitutionally acceptable level of scrutiny of the requesting state's evidence, and not to protect trial fairness in the requesting state. In the words of Ferras, at para. 47, the extradition judge scrutinizes the evidence to ensure that the appellant is not deprived of his "constitutional right to a meaningful judicial determination before [he] is sent out of the country and loses his . . . liberty" (emphasis in original).

[107] That is not to say Mohan and Abbey are of no use in the extradition context. While the inquiry in Mohan and Abbey is undertaken for a different purpose, and Mohan/Abbey and manifest unreliability analyses may therefore yield different results, the factors discussed in Abbey may nonetheless assist an extradition judge in determining whether an expert report in a record of the case is manifestly unreliable.

[108] This case illustrates the point. While the extradition judge declined to conduct a full Mohan/Abbey analysis, he nonetheless considered Mme Bisotti's qualifications and allowed the appellant to lead evidence questioning her qualifications and raising the issue of bias and the validity of her methodology. As Cronk J.A. pointed out in her decision granting the appellant bail pending appeal, the criteria set out in Mohan and Abbey are "general interpretive criteria applicable to the assessment [page202] of whether expert evidence meets the test of threshold reliability": unreported, July 19, 2011, at para. 34. Flexibly applied, these criteria may be useful in the extradition context.

[109] We would add one further observation. The appellant's argument assumes that a trial judge applying Mohan and Abbey would inevitably have kept the Bisotti report from the jury. We are not convinced that such is the case. According to the Goudge report, at pp. 488-89, a consideration in determining threshold reliability is whether other experts are available to evaluate the expert's technique. This allows the jury to understand the frailties in the tendered opinion. Applied to this case, the soft science of handwriting comparison is something much more understandable to a lay person than most matters upon which experts opine. Given the nature of the issue and the existence of competing expert reports, it is possible that a trial judge properly applying the principles in Mohan and Abbey might have used his or her discretion to admit the Bisotti report.

(2) Misapprehension of the evidence

[110] We are not persuaded that the extradition judge's conclusion that the Bisotti report was not manifestly unreliable was rooted in impermissible speculation and a misapprehension of the evidence.

[111] We agree with the appellant that the evidence of his experts was that there is one accepted method of handwriting analysis used around the world and that ENFHEX essentially endorses this universal method.

[112] ENFHEX has published a guide entitled "Overview Procedure for Handwriting Comparisons". It is an eight-page document that, in the words of the appellant's expert, Robert Radley, is "a very basic outline of general procedures together with some recommendations for reading material". The recommended works -- described in the guide as "some of the significant publications that relate to the examination of Handwriting" -- are published in the U.S., England and Germany. None of the referenced works is written in French.

[113] In her report, Mme Bisotti indicates that she followed the ENFHEX method and provides some detail of the methodology she employed. The appellant has not established that Mme Bisotti failed to follow the general procedures outlined in the ENFHEX guide. For example, the guide indicates that "[t]he approach relies on an examination of the characteristics of the handwriting, and an assessment of the similarities and differences found between pieces of Handwriting". The guide does not indicate that it is the differences that are determinative in [page203] establishing identification. And while the guide indicates that the expert should assess whether there is sufficient handwriting to be able to assess the range of variation, it does not direct what amount of handwriting constitutes a sufficient basis. The appellant's three experts -- none of whom was familiar with forensic document examination in France -- go into much more detail in their reports about how they say a competent handwriting expert would apply the general principles in the ENFHEX guide.

[114] Comparing the very general ENFHEX guide to the very detailed approach outlined by the appellant's experts, it would not have been unreasonable for the extradition judge at least to contemplate the possibility that forensic document examiners in France might have a different approach in their application of the ENFHEX guide in handwriting analysis.

[115] In any event, as we explained above, when the extradition judge denied the appellant's request for an adjournment to lead the evidence of Prof. Margot and Dr. Marquis, he made clear that evidence that France does not have a different approach to handwriting analysis would not have changed his decision that the Bisotti report was not manifestly unreliable. His conclusion was not rooted in the possibility that French forensic document examiners might use a different methodology.

(3) The committal judge imposed too high a standard

[116] As the extradition judge noted, a high threshold attaches to the categorization of evidence that may be called "manifestly unreliable". We are not persuaded that the extradition judge erred in the application of the Ferras standard. We see no basis for appellate intervention with his conclusion that the Bisotti report was not manifestly unreliable. In our view, the appellant has received the meaningful judicial determination that Ferras requires before he can be extradited.

[117] As noted above, Anderson did not deal with opinion evidence. Justice Doherty's remarks responded to the submission of the appellant in that case that Ferras entitled him to engage in what was effectively a speculative, unfocused credibility contest between two alleged co-conspirators. Justice Doherty's remark, at para. 28, that "Ferras, supra, does not envision weighing competing inferences that may arise from the evidence" should not be read as prohibiting an extradition judge from engaging in a limited weighing of an expert opinion in the record of the case, in light of competing expert opinions proffered by the person sought to determine whether it is manifestly unreliable. [page204]

[118] Expert opinions are admitted because they are beyond the scope of the expertise of the trier of fact. Because of this, in many cases it will be difficult for an extradition judge to conclude that the methodology used by the requesting state's expert is so clearly and obviously improper that the opinion is devoid of reliability and should be completely rejected without the assistance of other experts.

[119] Here, the extradition judge evaluated the expert opinion in the record of the case in light of the competing expert opinions proffered by the appellant. He carefully considered the evidence of the appellant's experts and the Bisotti report over the course of a protracted proceeding. He determined that the appellant's expert evidence did not lead him to conclude that the Bisotti report should be completely rejected as "manifestly unreliable".

[120] In Abbey, Doherty J.A. explained the discretionary nature of the gatekeeper analysis undertaken in deciding whether an opinion is sufficiently reliable to be considered by a jury. To repeat para. 79 of Abbey, quoted earlier, "[d]ifferent trial judges, properly applying the relevant principles in the exercise of their discretion, could in some situations come to different conclusions on admissibility". That statement is equally applicable to an extradition judge determining whether an expert opinion in the record of the case is manifestly unreliable.

[121] The appellant argues that the standard for manifest unreliability of an expert opinion reflected in para. 124 of the extradition judge's reasons is too high. For ease of reference, that paragraph, excerpted above, reads as follows:

If the only available conclusion derived from the [record of the case] and the evidence from the three experts presented on behalf of the person sought was that the French expert was biased, unqualified, and used improper methodology in every respect, then a finding of manifest unreliability would be possible. However, the evidence heard at this hearing did not support such an unequivocal finding.

[122] We agree with the appellant that para. 124, if read literally, and in isolation, reflects too high a standard. It is difficult to articulate where the line should be drawn, and the extradition judge struggled to do so. However, as we explain below, in our view it is clear from the extradition judge's reasons read as a whole that he understood and applied the correct standard.

[123] First, at para. 116, excerpted above, the extradition judge correctly wrote: "[M]anifest unreliability in the context of an extradition case would mean evidence that is so devoid of reliability that it should be rejected out of hand when considering the issue of committal". [page205]

[124] Second, it is clear from the example the extradition judge provided in para. 125 of what would amount to manifest reliability, that the references in para. 124 to bias, lack of qualification and improper methodology should be read disjunctively and not conjunctively. The example in para. 125 is with respect to methodology alone: the use by handwriting experts of exemplars not written by the appellant would render the opinion manifestly unreliable.

[125] Third, while the extradition judge wrote, in para. 124, that the competing opinions must show that the expert opinion in the record of the case used "improper methodology in every respect" (emphasis added), it is similarly clear from the example he provides at para. 125 that he appreciated that one methodological flaw -- if sufficiently fundamental -- can render an opinion manifestly unreliable.

[126] In our view, the standard effectively applied by the extradition judge was this: where the evidence of the expert whose opinion is in the record of the case is so suspect that it is devoid of reliability and of utility to the fact-finder, an extradition judge may properly find the opinion manifestly unreliable. And an extradition judge may come to that conclusion for one or more of a variety of reasons. This is, in our view, the correct standard where the opinion proffered is in a field of science or "pseudoscience" recognized in Canada as an admissible form of expert evidence, as is the case here.

(4) The test for committal

[127] In our view, the extradition judge correctly stated the test for committal, and, having concluded that the Bisotti report was not manifestly unreliable, did not err by not analyzing the evidence further to decide if it would be dangerous or unsafe to convict on all of the evidence in the record of the case. As we will explain, the "dangerous or unsafe to convict" standard properly belongs in the test for manifest unreliability, where the extradition judge applied it. Additionally, in our view the test for unreasonable verdict/ conviction is, by its very nature, different from the test for committal. The former is a retrospective analysis, performed once all of the evidence has been heard and weighed at the trial stage, while the latter is prospective, necessarily avoiding drawing conclusions on the ultimate outcome of the trial.

[128] In Ferras, the Supreme Court held that it is inappropriate to equate the task of the extradition judge with the task of a judge on a preliminary inquiry; it is not enough for evidence merely to exist on each element of the crime. The test applied on [page206] a preliminary inquiry could deprive the person sought of his constitutional right to a meaningful judicial determination before he is sent out of the country and loses his liberty. Rather, the evidence in the record of the case must demonstrably be able to be used by a reasonable, properly instructed jury to reach a guilty verdict. As explained above, this requires that the extradition judge engage in limited weighing of the evidence.

[129] As this court has previously held in Thomlison, Anderson and Michaelov, at the first stage, that limited weighing involves determining whether evidence in the record of the case is manifestly unreliable. If it is, it must not be considered by the extradition judge in stage two of the analysis in determining whether there is sufficient evidence to commit the person sought. This second stage requires the extradition judge to ask whether there remains "available and reliable" evidence upon which a reasonable jury, properly instructed, could convict. This stage will also involve limited weighing, to the extent that it is inevitable in a reasonableness assessment, where circumstantial evidence is involved.

[130] As the Supreme Court explained in R. v. Arcuri, [2001] 2 S.C.R. 828, [2001] S.C.J. No. 52, 2001 SCC 54, at para. 23, determining whether elements of an offence may reasonably be inferred from circumstantial evidence "inevitably requires the judge to engage in a limited weighing of the evidence because . . . there is . . . an inferential gap between the evidence and the matter to be established -- that is, an inferential gap beyond the question of whether the evidence should be believed" (citations omitted). As explained earlier in these reasons, this is exactly the test applied by the extradition judge, articulated at paras. 139-41.

[131] The appellant challenges this court's approach by focusing on the statement of the Supreme Court, at para. 54 of Ferras: "If the evidence is so defective or appears so unreliable that the judge concludes it would be dangerous or unsafe to convict, then the case should not go to a jury and is therefore not sufficient to meet the test for committal." He infers from this that the proper standard for the extradition judge to apply is the "unreasonable verdict/conviction" test.

[132] Shephard is the apparent source of the phrases "dangerous or unsafe to convict" and "manifestly unreliable" employed in Ferras. As mentioned above, in Shephard, a pre-Charter case, the majority of the Supreme Court held that an extradition judge could not decline to commit the person sought on the basis that the evidence proffered by the requesting state was manifestly unreliable. Shephard seemingly uses the phrases [page207] "manifestly unreliable" evidence and evidence that is so dubious as to make it "dangerous or unsafe to convict" interchangeably. In other words, it would be dangerous or unsafe to convict if the evidence is manifestly unreliable. As discussed above, in Ferras, the Supreme Court essentially adopts the dissenting reasons of Spence J. in Shephard.

[133] It is clear from the extradition judge's oral reasons of February 18, 2011 [[2011] O.J. No. 2398 (S.C.J.)] that he addressed whether it would be dangerous to convict based on the Bisotti report. There, he phrased [at para. 4] the issue before him as whether the Bisotti report "should be excluded from the Record of the Case as manifestly unreliable, or as so defective that it would be dangerous to convict upon [it ]".

[134] In our view, the "dangerous or unsafe to convict" consideration in Ferras is addressed by the exclusion of evidence that is manifestly unreliable. The test for committal is made out where (1) evidence that is manifestly unreliable, and on which it would therefore be dangerous or unsafe to convict, is excluded; and (2) there remains evidence available for trial that possesses indicia of threshold reliability on each essential element of the corresponding Canadian offence, upon which a reasonable jury properly instructed could convict.

[135] We reject the appellant's argument that the extradition judge must apply a test comparable to the "unreasonable verdict/conviction" test, and the suggestion that such test requires an appellate court to ask whether the verdict was unsafe. We disagree with the British Columbia Court of Appeal in Graham to the extent that it holds otherwise.

[136] In Graham, decided six months after Thomlinson and Anderson were released, Donald J.A. criticized what he regarded as this court's emphasis on the concept of "manifestly unreliable" evidence as a component of the test for committal. Quoting para. 28, excerpted above, of Doherty J.A.'s decision in Anderson to the effect that an extradition judge has no authority to deny committal in a case that appears "weak" or "unlikely to succeed at trial", Donald J.A. disagreed and continued, at para. 31:

My difficulty with that way of putting the limit of discretion is that there are degrees of weakness and extradition judges should not be put off their task of assessing for sufficiency by a dictum that mere weakness is not enough. I agree that it is not enough. It may be more helpful to speak in terms of the test for reviewing verdicts for unreasonableness on appeal or for directed verdicts.

(Emphasis in original)

[137] The unreasonable verdict test was stated by the Supreme Court in Yebes, at p. 185 S.C.R.: [page208]

[C]urial review is invited whenever a jury goes beyond a reasonable standard. In my view, then . . . the test is "whether the verdict is one that a properly instructed jury acting judicially, could reasonably have rendered".

(Emphasis added)

[138] As the Supreme Court explained in R. v. Biniaris, [2000] 1 S.C.R. 381, [2000] S.C.J. No. 16, 2000 SCC 15, at para. 36:

[The] formulation of the test imports both an objective assessment and, to some extent, a subjective one. It requires the appeal court to determine what verdict a reasonable jury, properly instructed, could judicially have arrived at, and, in doing so, to review, analyse and, within the limits of appellate disadvantage, weigh the evidence.

(Emphasis added)

And, at para. 40:

[A]cting judicially means not only acting dispassionately, applying the law and adjudicating on the basis of the record and nothing else. It means, in addition, arriving at a conclusion that does not conflict with the bulk of judicial experience.

[139] The unreasonable verdict test is applied after trial, on a full record, and permits the appeal court to weigh the evidence, subject only to "the limits of appellate disadvantage": in other words, it is a retrospective analysis. This is a completely different perspective from that contemplated by s. 29(1) of the Act, which speaks of evidence that "would justify committal for trial in Canada". In Ferras, at para. 46, the Supreme Court made clear that the test on an extradition hearing is whether the requesting state has established "a case that could go to trial in Canada" (emphasis in original). The analysis is prospective. And, as discussed above, the Supreme Court also made clear in Ferras that an extradition judge is only to engage in a limited weighing of the evidence.

[140] Further, in R. v. G. (A.), [2000] 1 S.C.R. 439, [2000] S.C.J. No. 18, 2000 SCC 17, which was heard together with Biniaris, the Supreme Court rejected a formulation of the unreasonable verdict test that inquired whether the verdict was "unsafe" and confirmed its formulation of the unreasonable verdict test in Biniaris. The existence of "dangerous or unsafe to convict" language in the test for committal, explored above, does not support the use of the unreasonable verdict standard. The formulations of the tests for unreasonable verdict and for committal are thus quite different, although they may seem similar on the surface.

[141] For these reasons, the committal appeal is dismissed. We now turn to the appellant's application for judicial review of the Minister's extradition order. [page209]

IV. The Judicial Review Application

[142] On the judicial review application, the appellant advances two grounds: (1) the Minister lacked the jurisdiction to order surrender because the appellant is not wanted by France for the purpose of a trial; and (2) to surrender the appellant to face a trial based (a) on unknown, unsourced and uncircumstanced intelligence reports, and (b) on evidence contained in those reports that may have been derived from the use of torture, is contrary to s. 7 of the Charter as it would shock the conscience of Canadians.

[143] We will deal with each of these grounds in turn.

A. Did the Minister have jurisdiction?

[144] The appellant argues that the Minister erred in concluding that he had jurisdiction to entertain France's extradition request. In the appellant's submission, the Act, properly interpreted, limits the Minister's authority to order extradition to cases where the requesting state has decided to put the person on trial. Because it is acknowledged that France has yet to make a decision as to whether the appellant will be put on trial, the Minister acted without jurisdiction and his order ought to be quashed.

[145] We would not give effect to this submission.

(1) The statutory authority

[146] Extradition from Canada is governed by the Act, which sets out the Minister's power to surrender. The appellant, in his submissions to the Minister, explained that the Minister's power is circumscribed and its reasonableness must be assessed in the context of the Act. As explained by the appellant:

1. The Minister's power to order or refuse surrender is subject to the provisions of the treaty and the Act and must be exercised in accordance with the Canadian Charter of Rights and Freedoms.

2. The reasonableness of the [Minister's] decision must be assessed with regard to all relevant circumstances and the applicable provisions of the Act.

3. Any inconsistency between the Act and the relevant treaty is to be resolved in favour of and governed by the Act: "For greater certainty, if there is an inconsistency between this Act and a specific agreement, this Act prevails to the extent of the inconsistency."

(Citations omitted)

[147] The appellant's argument that the Minister did not have the authority to grant France's extradition request is based on s. 3(1) of the Act. This section outlines the general principle of [page210] the Act. For our purposes, the critical portion of s. 3(1) is the introductory paragraph, which reads as follows:

3(1) A person may be extradited from Canada in accordance with this Act and a relevant extradition agreement on the request of an extradition partner for the purpose of prosecuting the person or imposing a sentence on -- or enforcing a sentence imposed on -- the person if . . .

(Emphasis added)

[148] In the appellant's submission, s. 3 of the Act imposes at least two limits on jurisdiction. First, the request must come from an extradition partner, that is, a country with whom an extradition agreement has been entered into. It is clear, therefore, that the Minister has no authority to extradite a person unless the request comes from an extradition partner.

[149] A second limit is imposed on the Minister: the surrender request must be for "the purpose of prosecuting the person" or, in the French version, "pour subir son procès".

(2) The standard of review

[150] Judicial review to this court is pursuant to s. 57 of the Act. Section 57(7) provides that this court can grant relief from the Minister's decision on any of the grounds set out in s. 18.1(4) of the Federal Courts Act, R.S.C. 1985, c. F-7. Included in those grounds is whether the Minister acted without jurisdiction or erred in law. As the Supreme Court stated in Lake v. Canada (Minister of Justice), [2008] 1 S.C.R. 761, [2008] S.C.J. No. 23, 2008 SCC 23, at para. 26: "[U]nder s. 57(2) [of the Act], judicial assessment of the Minister's decision by the court of appeal is a form of administrative law review and must be conducted in accordance with the applicable administrative law standard."

[151] In the appellant's submission, the Minister erred in concluding that he had the necessary jurisdiction, and correctness is the applicable standard for reviewing this conclusion. The appellant explains that because extradition is wholly a creature of statute, the Minister's power must be anchored in the statute: see McVey (Re); McVey v. United States of America, 1992 48 (SCC), [1992] 3 S.C.R. 475, [1992] S.C.J. No. 95, at p. 508 S.C.R. Faced with an extradition request, the Minister's jurisdiction under the Act is only triggered where (1) the request comes from an extradition partner and (2) the request is for the purpose of prosecuting the person sought. The existence of each of these elements is therefore, in the appellant's submission, a "true question of jurisdiction" attracting review on a correctness standard.

[152] We disagree. When the Supreme Court reformulated the standard of review analysis for judicial review in Dunsmuir v. New Brunswick, [2008] 1 S.C.R. 190, [2008] S.C.J. No. 9, 2008 SCC 9, [page211] it retained "true questions of jurisdiction" as a category attracting correctness review: see, e.g., para. 59. However, the court has since sharply called into question the viability of the "true questions of jurisdiction" category, notably in Alberta (Information and Privacy Commissioner) v. Alberta Teachers' Assn., [2011] 3 S.C.R. 654, [2011] S.C.J. No. 61, 2011 SCC 61 -- the very decision the appellant cites in support of his claim that the language quoted above from s. 3(1) of the Act presents a true question of jurisdiction. Writing for a six-member majority of the court, Rothstein J. suggested, at para. 34, that "it may be that the time has come to reconsider whether, for purposes of judicial review, the category of true questions of jurisdiction exists and is necessary to identifying the appropriate standard of review".

[153] Justice Rothstein's reasons in Alberta Teachers' Assn. did not explicitly eliminate the category of true questions of jurisdiction, but given that he could not reference a single legitimate application of the category, he was "unable to provide a definition of what might constitute a true question of jurisdiction": at para. 42. He held, at para. 39: "As long as the true question of jurisdiction category remains, the party seeking to invoke it must be required to demonstrate why the court should not review a tribunal's interpretation of its home statute on the deferential standard of reasonableness."

[154] This burden will rarely, if ever, be met: "Experience has shown that the category of true questions of jurisdiction is narrow indeed. Since Dunsmuir, this court has not identified a single true question of jurisdiction": Alberta Teachers' Assn., at para. 33. Citing Alberta Teachers' Assn., this court has found that "true questions of jurisdiction will be 'exceptional'": Pastore v. Aviva Canada Inc. (2012), 112 O.R. (3d) 523, [2012] O.J. No. 4508, 2012 ONCA 642.

[155] Granted there is a complication in this case because, as noted above, whether the Minister acted without jurisdiction is listed in s. 18.1(4) of the Federal Courts Act as one ground on which this court can grant relief from the Minister's decision. However, that the court can grant relief on this ground is not disputed -- the proper standard for judicial review is. Further, as also noted above, this exercise of judicial review "is a form of administrative law review and must be conducted in accordance with the applicable administrative law standard": Lake, at para. 26. Indeed, in Lake, the Supreme Court addressed the standard of review generally applicable to the Minister's decision to surrender an individual to the requesting state. Justice LeBel, writing for a unanimous court, stated, at para. 22: [page212]

After an individual has been committed for extradition, the Minister reviews the case to determine whether the individual should be surrendered to the requesting state. This stage of the process has been characterized as falling "at the extreme legislative end of the continuum of administrative decision-making" and is viewed as being largely political in nature: Idziak v. Canada (Minister of Justice), 1992 51 (SCC), [1992] 3 S.C.R. 631, at p. 659.

[Emphasis in original]

[156] Justice LeBel noted further, at para. 34, that the Supreme Court "has repeatedly affirmed that deference is owed to the Minister's decision whether to order surrender once a fugitive has been committed for extradition", and that reasonableness is the appropriate standard of review even where "the fugitive argues that extradition would infringe his or her rights under the Charter". If deference is owed to the Minister's interpretation of Charter provisions in the context of a decision to surrender, then, in our view, deference is a fortiori owed to the Minister's interpretation of provisions such as s. 3(1) of the Act -- with which he is especially familiar, and in the application of whose provisions he has particular expertise -- in the context of such a decision.

[157] This result is confirmed by the approach Rothstein J. adopted in Alberta Teachers' Assn., at para. 44: "Dunsmuir provided guidance as to how a standard of review might be determined summarily without requiring a full standard of review analysis. One method was to identify the nature of the question at issue, which would normally or, I say, presumptively determine the standard of review." On this approach, Rothstein J. concluded, at para. 46, that "since Dunsmuir, for the correctness standard to apply, the question has to not only be one of central importance to the legal system but also outside the adjudicator's specialized area of expertise" (emphasis in original).

[158] Given Lake's characterization of the Minister's decision as "being largely political in nature", the nature of this decision is clearly not of central importance to the legal system or outside the Minister's area of expertise. The approach adopted in Alberta Teachers' Association thus confirms the deference owed to the Minister's decision on judicial review.

[159] A reasonableness standard is therefore appropriate for review of the Minister's decision regarding the application of s. 3(1) of the Act to the particular circumstances of this case. The Minister's decision is one of mixed fact and law in that he interprets the Act and assesses how it applies to the factual circumstances of the extradition request. His decision is entitled to deference and ought not to be interfered with unless the appellant can show that it is unreasonable. As we will explain, the [page213] Minister's analysis and conclusion on the jurisdictional issue were reasonable.

(3) The French system

[160] Before turning to the legal issue raised, a brief description of the relevant aspects of the French criminal justice system is necessary.

[161] France has a civil law system. In criminal matters a juge d'instruction is assigned to investigate crimes and may place a specific individual under judicial investigation ("mis en examen"). Before an individual can be mis en examen, that individual must have made a first appearance before the juge d'instruction and have been allowed to make a statement. Once an individual is mis en examen, the individual officially becomes the subject of a criminal investigation.

[162] When the investigation is complete, the juge d'instruction notifies the parties and decides whether the evidence justifies referring the matter to a trial court or declaring that there is no case to answer.

[163] With respect to this case, the French government has confirmed that the appellant has not yet been mis en examen. Once he appears before the juge d'instruction, the juge d'instruction will decide whether he will be mis en examen and, assuming he is mis en examen, only then can the judge complete the investigation and decide whether the appellant will stand trial. As explained by the appellant, it is clear, therefore, that no decision has been made to have the appellant tried in France.

[164] France has explained why the appellant cannot be mis en examen and sent to trial until after the appellant has appeared in person before the juge d'instruction. This is to ensure that the juge d'instruction has the opportunity to hear from the appellant and allow him to comment on the evidence that has been assembled before the juge d'instruction moves to the next phase. As we will explain, the appellant challenges France's assertion that his appearance is necessary before a decision is made as to whether he will be mis en examen.

(4) What does "for the purpose of prosecuting the person" -- "pour subir son procès" -- mean?

[165] It is well settled that the Act does not allow the extradition of a person for mere investigative purposes. Extradition is not to be used as a tool by foreign states to question people as potential witnesses or suspects. To trigger application of the Act, more is required. [page214]

[166] The present case requires us to consider more closely the threshold that must be met in order to satisfy the requirement in s. 3(1) that the extradition request be for "the purpose of prosecuting the person".

[167] In our view, although extradition is grounded on the anticipation of a foreign trial, it need not be inevitable that a trial will occur. Different legal systems will have different processes and steps required to be taken before a trial takes place. In Canada, for example, a person charged may be subjected to a preliminary inquiry before proceeding to trial. The person charged will not, therefore, necessarily undergo a trial.

[168] Different systems will also have different safeguards in place to ensure fairness and due process. Again, by way of example, in Canada a prosecutor exercises prosecutorial discretion and, based on information received right up to and including the trial, can decide to discontinue the prosecution. Similarly, in the course of pre-trial proceedings, a judge can, in appropriate circumstances, stay the charge and thereby obviate the need for a trial.

[169] It is accepted that the differences in processes and procedures are greatest when one compares a common law system such as ours with a civil law system such as France's.

[170] In Ismail (Re), [1998] UKHL 32, [1999] 1 A.C. 320 (H.L.), the House of Lords, dealing with an extradition request from the Federal Republic of Germany, was presented with the argument that because no formal criminal charge had been laid by Germany against the person sought, the person sought was not an "accused" as required by the Extradition Act 1989 (UK), 1989, c. 33. Lord Steyn, writing for the court, rejected this submission. He cautioned that a broad and purposive interpretation was required. In reaching this conclusion, he made several points that apply equally to the Canadian context, including:

"mere suspicion that an individual has committed an offence is insufficient to place him in the category of 'accused' persons. . . . Something more is required": at p. 326 A.C.;

"one is concerned with the contextual meaning of 'accused' in a statute intended to serve the purpose of bringing to justice those accused of serious crimes. There is a transnational interest in the achievement of this aim": at p. 326-27 A.C.;

"it would be wrong to approach the problem of construction [of 'accused'] from the perspective of [domestic] criminal procedure" given the divergent systems of law involved and notably the differences between common law and civil law jurisdictions: at p. 327 A.C.; [page215]

a "broad" and "purposive interpretation of 'accused' ought to be adopted in order to accommodate the differences between legal systems". The approach should be "cosmopolitan" and focus on matters of substance over form: at p. 327 A.C.; and

the test is "whether the competent authorities in the foreign jurisdiction ha[ve] taken a step which can fairly be described as the commencement of a prosecution": at p. 327 A.C.

[171] Although these points were made with respect to the U.K.'s extradition statute, which speaks in terms of an "accused", the same principles apply, in our view, to the Act when it refers to extradition "for the purpose of prosecuting the person".

[172] The appellant argues that the approach adopted in Ismail should not be followed and the approach taken in Fletcher v. Government of France (2008), 2007-09 Gib. L.R. 191 (S.C. Gibraltar), and The Minister for Justice, Equality and Law Reform v. Bailey, [2012] IESC 16 (S.C. Ireland) is to be preferred.

[173] We do not view the Fletcher and Bailey cases as providing useful guidance. In both cases, the court was considering an extradition request made pursuant to a European arrest warrant. Extradition in the Fletcher case was only authorized by the applicable statute where the requesting state certified that both a proceeding against the person had been commenced and "a decision to try him for the offence concerned had been made" (emphasis added): Fletcher, at para. 6. The statute in the Bailey case was even stronger. It provided that the court "shall refuse to surrender the person if it is satisfied that a decision has not been made to charge the person with, and try him or her for, that offence in the issuing state" (emphasis added): Bailey, at para. 70.

[174] The wording of the provisions in those two cases made it quite clear that for extradition to occur, both a decision to initiate a proceeding and a decision to try the person sought had to have been made. There is no such requirement under our legislation. Our legislation is more in line with the British legislation and the approach taken in Ismail is to be preferred.

[175] What we conclude from the cases and a contextual reading of the statutory provisions is that an extradition request must be grounded in the anticipation that there will be a foreign trial. In other words, a process or prosecution has been initiated against the person sought that will, if not discontinued, lead to a trial. The person sought must be more than a mere suspect. A trial of that person, however, need not be inevitable.

[176] The record in this case clearly demonstrates that the appellant, if extradited, will not simply "languish in prison": see [page216] Ferras, at para. 55. The appellant is more than a mere suspect in that France has taken steps that can fairly be described as the commencement of a prosecution against him. Specifically, the record discloses that the juge d'instruction has issued an arrest warrant for the appellant. The juge d'instruction is satisfied that the French authorities obtained "very specific information" of the appellant's alleged participation in the French offences that has been "corroborated by [subsequent] investigations". When the appellant is surrendered, he will be taken into custody and brought before the juge d'instruction for his first appearance and, at that appearance, he will be given an opportunity to be heard. A further indication that a trial is anticipated flows from France certifying in the record of the case that the evidence set out in the record of the case is available for trial.

[177] In our view, the Minister reasonably concluded that the French authorities have taken steps consistent with the commencement of a prosecution against the appellant. The French arrest warrant clearly shows that the appellant is an accused person charged with an offence and wanted for prosecution. He is much more than a mere suspect or someone wanted for questioning.

(5) Is the Minister's interpretation of s. 3 inconsistent with the French version of the statute?

[178] The appellant argues that the French version of the statute is more demanding than the English version. The French version provides that the extradition request must be for the purpose of having the person sought "subir son procès", that is, to stand trial. In the appellant's submission, although the English version "for the purpose of prosecution" is broader and might not require that a decision have been taken that the person sought will be required to stand trial, the French version makes it clear that the purpose of the request must be to have the person stand trial. Because the two language versions of s. 3 differ, the rules of bilingual interpretation are to be applied to determine the proper meaning and scope of s. 3.

[179] The appellant explains that where one version is clear and the other is ambiguous but reasonably capable of bearing the meaning of the clear version, the one expressed by the clear provision is to be preferred: see, e.g., Schreiber v. Canada (Attorney General), [2002] 3 S.C.R. 269, [2002] S.C.J. No. 63, 2002 SCC 62, at para. 56. Similarly, where one version is narrow, and the other is broad but can reasonably be read to include the narrower version, the preferred meaning is that expressed by the narrow version: see Pierre-André Côté, Stéphane Beaulac and [page217] Mathieu Devinat, The Interpretation of Legislation in Canada, 4th ed. (Toronto: Carswell, 2011), at pp. 347-49.

[180] In the appellant's submission, when both versions are compared, the English version that refers to "prosecution" can reasonably be read to share the meaning of the clearer and narrower French version that speaks of "trial". The tenets of bilingual interpretation of statutes therefore dictate that the correct interpretation is the shared meaning as expressed in the French version, which is that the person sought be sought to stand trial.

[181] The appellant notes that this conclusion is reinforced when reference is made to s. 33(3) of the Act. That section provides that a judicial or prosecuting authority of the requesting state must certify that the evidence summarized in the record of the case "is available for trial".

[182] The appellant acknowledges that the Minister referred to the French version but submits that the Minister did not give effect to its clear meaning: that the requesting state must have made a decision to send the person sought to trial.

[183] In our view, consideration of the French version does not lead to the interpretation advanced by the appellant. We acknowledge that, standing alone, the French version of s. 3(1) could be read as requiring that the requesting state has made the decision to try the person sought. The reference to standing trial in the French version is more specific than the English version that refers to the person being sought for prosecution. Even though it is more specific than the English version, the French version does not go as far as suggested by the appellant. It does not provide that a decision must have been made by the requesting state that the person is to stand trial. It simply provides that the request is made so that he can stand trial in the requesting state.

[184] The correct meaning, in any case, should not be based on an interpretation of the words of s. 3(1) read in isolation. The modern principle of statutory interpretation is that "the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament": Rizzo and Rizzo Shoes Ltd. (Re) (1998), 1998 837 (SCC), 36 O.R. (3d) 418, [1998] 1 S.C.R. 27, [1998] S.C.J. No. 2, at para. 21 (quoting Elmer Driedger, Construction of Statutes, 2nd ed. (Toronto: Butterworths, 1983)).

[185] We agree that the principles of interpretation of bilingual statutes provide that where one version has a broader meaning than the other, the shared meaning, the meaning to be adopted, will usually be the narrower of the two. However, statutory interpretation involves more than these principles. As [page218] explained by Binnie J. in Canada (Citizenship and Immigration) v. Khosa, [2009] 1 S.C.R. 339, [2009] S.C.J. No. 12, 2009 SCC 12, at para. 39:

Linguistic analysis of the text is the servant, not the master, in the task of ascertaining Parliamentary intention . . . A blinkered focus on the textual variations [between the English and French versions] might lead to an interpretation at odds with the modern rule because, standing alone, linguistic considerations ought not to elevate an argument about text above the relevant context, purpose and objectives of the legislative scheme[.]

(Citations omitted)

[186] In our case, the French authorities advise that the appellant's personal presence before the juge d'instruction is required before he can be mis en examen and referred to trial. To interpret the Act as the appellant suggests would put Canada in breach of its obligations under the extradition treaty with France and, likely, under extradition treaties entered into with other countries that are civil law jurisdictions and have juges d'instructions who operate in similar regimes. Such an interpretation ought to be avoided: see National Corn Growers Assn. v. Canada (Import Tribunal), 1990 49 (SCC), [1990] 2 S.C.R. 1324, [1990] S.C.J. No. 110, at pp. 1371-1372 S.C.R.; R. v. Hape, [2007] 2 S.C.R. 292, [2007] S.C.J. No. 26, 2007 SCC 26, at para. 53.

[187] Although we agree with the appellant that the Act does not specifically implement the treaty, the Act is intended to empower the Minister to give effect to the various treaties entered into by the government. The Minister's authority, therefore, comes from the Act but the Act cannot be interpreted in a vacuum. As explained in Hape, at para. 53, there is a well-established principle of statutory interpretation that "as a matter of law, courts will strive to avoid constructions of domestic law pursuant to which the state would be in violation of its international obligations, unless the wording of the statute clearly compels that result"; see, also, United States of America v. Anekwu, [2009] 3 S.C.R. 3, [2009] S.C.J. No. 41, 2009 SCC 41, at para. 25.

[188] The U.K. courts recently rejected a similar argument in Assange v. Swedish Prosecution Authority, [2011] EWHC 2849 (Admin), affd [2012] UKSC 22. Mr. Assange, who was the subject of an arrest warrant for various offences in Sweden, argued that he was not extraditable to Sweden, a civil law country, because the case against him was only at the investigative stage. Under Swedish law, he could not be formally charged until he had appeared before the investigating judge and been questioned. It was acknowledged that further pre-trial investigation might result in no trial ever taking place. Mr. Assange maintained that in these circumstances, he did not have the status of an [page219] "accused" wanted for prosecution as required by the British statute. In rejecting this argument, the High Court of Justice applied the reasoning in Ismail and concluded that the fact that additional pre-trial investigation in Sweden might result in no trial taking place did not mean that Mr. Assange was a suspect as opposed to an accused.

[189] In the present case, the Minister also relied on the reasoning in Ismail. As explained by the Minister:

[T]he extradition process mandates an approach that is not constrained by domestic notions of what constitutes an accused or charged person, or the end of the investigative stage and the commencement of the trial stage in a given case (Re Ismail, [1999] 1 A.C. 320 (H.L.); United States of America v. Coffey (2006), 210 C.C.C. (3d) (Man. C.A.); Canada v. Schmidt, 1987 48 (SCC), [1987] 1 S.C.R. 500).

Rather, in the interest of accommodating differences between legal systems, I am required to take an approach that interprets both the Treaty and the Act in a purposive and flexible manner. To that end, I must not subject the judicial process in France to overly technical evaluations against the rules that govern the legal process in Canada (Canada v. Schmidt, supra).

[190] In our view, the conclusion reached by the Minister is in keeping with the context and objectives of the Act and is consistent with both the French and English versions of s. 3(1) of the Act.

(6) Did the Minister err in accepting the French authorities' interpretation of the procedure under French law?

[191] The appellant also argues that the Minister erred in relying on information provided by the French authorities to the effect that the appellant's personal presence before the juge d'instruction is required before a decision can be made to have the appellant mis en examen and refer the case to trial. The appellant submits that the Minister should have preferred the opinions of his experts, Mr. Stéphane Bonifassi and Prof. Jacqueline Hodgson, over the advice received by the Minister from the French authorities. Mr. Bonifassi is a criminal defence lawyer in France. Prof. Hodgson is a law professor with expertise in French criminal justice.

[192] Mr. Bonifassi expressed the view that the appellant is currently only a mere suspect. He asserted, contrary to the information provided by France, that the appellant does not have to appear before the juge d'instruction for his case to be referred to trial. While Mr. Bonifassi conceded that appearing before the judge d'instruction is "the usual way" for referring a person to trial, he maintained that there is an alternative method. Prof. Hodgson expressed a similar view. In her opinion, [page220] France was requesting the appellant's extradition not to stand trial but rather to continue the investigation of the case, a case in which he is a mere suspect.

[193] We disagree. The Minister reasonably relied on the information provided by the French authorities instead of the opinions of Mr. Bonifassi and Prof. Hodgson. The application of French law is a matter for the French authorities. While the Minister can take foreign law into account in deciding whether to order surrender, he does not make his own assessment of how foreign law should apply, as to do so would offend the principle of comity: see Canada (Justice) v. Fischbacher, [2009] 3 S.C.R. 170, [2009] S.C.J. No. 46, 2009 SCC 46, at para. 54; France v. Ouzghar (2009), 94 O.R. (3d) 601, [2009] O.J. No. 277, 2009 ONCA 69, at para. 16.

[194] As the Minister noted:

It is not for me to resolve any conflict that may exist regarding the interpretation and implementation of France's law and procedure. Absent evidence of bad faith or improper motive on the part of France, I am entitled and indeed obliged to rely on France's clarification of its legal system[.]

[195] The Minister refers to Fischbacher, among other authorities. In Fischbacher, at para. 5, Charron J., writing for an eight-member majority, notes that an approach requiring "the Minister to effectively second-guess the foreign state's assessment of its own law . . . finds no support in the Extradition Act and . . . offends the fundamental principle of comity"; see, also, para. 51.

[196] In any event, we do not understand the matter to be as straightforward as suggested by Mr. Bonifassi. He maintains that the combined operation of arts. 134, 175 and 176 of France's Code of Criminal Procedure provides that where an arrest warrant has been issued, but the individual cannot be arrested, that individual is deemed to have been mis en examen and may subsequently be referred to trial if the juge d'instruction believes that there is sufficient evidence to warrant it. The fact is, however, that the search for the appellant has not been fruitless. The appellant has now been arrested, albeit by the Canadian authorities. The effect of such an arrest on the application of arts. 134, 175 and 176 of the Code of Criminal Procedure is not as clear as Mr. Bonifassi seems to suggest.

[197] Further, even if the juge d'instruction has the discretion to refer the appellant's case to trial in absentia, the Minister recognized that it is not for Canadian authorities to resolve issues surrounding the application of French criminal law and procedure as part of extradition proceedings. As explained by the Minister, [page221]

[E]ven if the investigating judge has the discretion to close the case and refer Mr. Diab to trial without requiring his physical presence in France, I have no doubt that the French authorities have followed the appropriate and most sensible course of action in this case. France has explained that the investigating judge has a truth-seeking function and the requirement that the person under investigation be heard before placing them under judicial examination is an important part of discharging this role. Indeed, it largely appears to be a measure that protects the interests of the person who is the subject of an arrest warrant by ensuring that the case dossier is complete and includes the defence position. In the face of pending extradition proceedings, which may result in Mr. Diab's surrender to France, a decision to forego this requirement would seem to be unwise and untimely.

[198] In conclusion, we view the Minister's decision that he had jurisdiction to order the appellant's extradition to be reasonable.

B. Trial fairness issues (evidence based on intelligence reports and/or potentially tainted by torture)

[199] The appellant's second argument on the judicial review application is a Charter argument. He submits that his Charter rights would be infringed by his surrender to face trial in France because the evidence in the French proceeding may be based, at least partly, on unsourced and unknown intelligence reports and, more particularly, on evidence that may have been obtained by using torture.

(1) The test for refusing surrender and the standard of review

[200] Section 44(1)(a) of the Extradition Act provides that the Minister shall refuse to make a surrender order if satisfied that the surrender would be unjust or oppressive having regard to all the relevant circumstances.

[201] We note that on this ground of appeal, the Minister's decision is attacked under s. 7 of the Charter -- the right not to be deprived of life, liberty and security of the person except in accordance with the principles of fundamental justice. This may lessen the level of deference: see, e.g., United States of America v. Kwok, [2001] 1 S.C.R. 532, [2001] S.C.J. No. 19, 2001 SCC 18. However, review on a standard of reasonableness remains the norm even where "the fugitive argues that extradition would infringe his or her rights under the Charter": Lake, at para. 34.

[202] The authorities establish that, if surrender would violate s. 7 of the Charter, it will also bar surrender under s. 44(1)(a). However, the authorities also confirm that the test for refusing surrender on s. 7 grounds is a "strict one", and only precludes surrender in cases of a "very exceptional nature" [page222] where surrender to the requesting state would "shock the conscience" of Canadians and be "simply unacceptable": Kindler v. Canada (Minister of Justice), 1991 78 (SCC), [1991] 2 S.C.R. 779, [1991] S.C.J. No. 63, at pp. 849-50 S.C.R.; United States v. Burns, [2001] 1 S.C.R. 283, [2001] S.C.J. No. 8, 2001 SCC 7, at paras. 66-69; Canada v. Schmidt, 1987 48 (SCC), [1987] 1 S.C.R. 500, [1987] S.C.J. No. 24, at p. 522 S.C.R.; and Canada (Minister of Justice) v. Pacificador (2002), 2002 41595 (ON CA), 60 O.R. (3d) 685, [2002] O.J. No. 3024 (C.A.), at para. 47, leave to appeal to S.C.C. refused [2002] S.C.C.A. No. 390, 101 C.R.R. (2d) 374.

[203] Provided the Minister applies the correct legal test and does not otherwise err in law or contravene the principles of natural justice, his or her decision that surrender would not be contrary to s. 7 of the Charter and s. 44(1) (a) of the Extradition Act remains entitled to considerable deference.

[204] The appellant's Charter argument has two parts and we will address each part separately.

(2) Use of intelligence-based evidence

[205] The frailties of using evidence from international intelligence agencies are universally acknowledged. The source of the evidence is unknown. The circumstances in which the evidence was gathered are unknown. Often, the intelligence evidence itself is unknown because, for national security reasons, the named person is denied access to it. In the appellant's words, the intelligence information is "unsourced, uncircumstanced, and unknown". In his submissions to the Minister, the appellant argued that the case against him consisted of "bald and conclusory, anonymous intelligence assertions of his alleged guilt": reasons for surrender, at p. 9.³

[206] As the appellant asserts, the far-reaching implication of those frailties is that a named person -- whether in extradition/surrender proceedings, security certificate proceedings, non-refoulement or other similar proceedings -- has no meaningful opportunity to challenge, test and refute the information, either in the domestic proceeding or at the proposed trial in the requesting state. This raises concerns about the violation of deep-seated Canadian and international values relating to the principles of fundamental justice, not the least of which is the right to a fair trial, including the right of a person in jeopardy to know the case he or she is facing and to respond effectively to it. [page223]

[207] The concerns with intelligence-based information are reflected in Canadian and international law and in the reports of a number of non-governmental organizations and public inquiries: Charkaoui v. Canada (Citizenship and Immigration), [2007] 1 S.C.R. 350, [2007] S.C.J. No. 9, 2007 SCC 9, at para. 61; International Covenant on Civil and Political Rights, December 19, 1966, 999 U.N.T.S. 171, art. 14(3)(e); Dennis O'Connor, Report of the Events Relating to Maher Arar (Ottawa: Minister of Public Works and Government Services, 2006); Frank Iacobucci, Internal Inquiry Into the Actions of Canadian Officials in Relation to Abdullah Almalki, Ahmad Abou-Elmaati and Muayyed Nureddin (Ottawa: Minister of Public Works and Government Services, 2010); Human Rights Watch, Preempting Justice: Counterterrorism Laws and Procedures in France (New York: Human Rights Watch, 2008); Human Rights Watch, "No Questions Asked": Intelligence Cooperation with Countries that Torture (New York: Human Rights Watch, 2010); and International Commission of Jurists, Assessing Damage, Urging Action: Report of Eminent Jurists Panel on Terrorism, Counter-terrorism and Human Rights (Geneva: International Commission of Jurists, 2009).

[208] The appellant and the CCLA argue that it would be unacceptable, as a matter of principle, to extradite someone where there is the potential that unsourced intelligence information will be used against the person at the person's trial in the requesting state. Although they recognize the importance of Canada's international obligations based on comity and treaties, they submit that these considerations can never trump the Charter obligation to protect an individual from being sent abroad to face a situation that fails to accord with our most basic expectations of fairness and decency.

[209] We agree. However, the key consideration is the potential for that situation to arise. We do not think there should be a categorical exclusionary rule against resort to intelligence-based information in these kinds of situations. To impose such a rule would effectively eviscerate the ability of Canadian and international authorities to bring terrorists to justice because the evidence in such cases is very often sourced through international intelligence agencies. The central issue is the risk that such evidence will be used at trial against the named person in a fashion that fails to protect the person's fundamental right to make answer and defence and have the benefit of a fair trial.

[210] The appellant and the CCLA rely on the decision of the Supreme Court of Canada in Charkaoui. In that case, Mr. Charkaoui argued that the security certificate regime then in place under the [page224] Immigration and Refugee Protection Act, S.C. 2001, c. 27 was unconstitutional because it contravened the s. 7 Charter rights of persons facing detention and deportation for terrorism-related reasons. The court agreed.

[211] Central to the court's conclusion was the view that the security certificate system, as then designed, provided no adequate alternative safeguards and unnecessarily compromised the right to a fair hearing by denying the person named in the certificate the opportunity to know the case put against him or her and, therefore, deprived that person of the ability to challenge the government's case. Interestingly -- given the appellant's attack on the role of the French juges d'instruction here -- one of the factors taken into account by the court in finding the security certificate system wanting was that Canadian judges, as the system was then structured, were "not afforded the power to independently investigate all relevant facts that true inquisitorial judges enjoy": Charkaoui, at para. 51.

[212] Charkaoui affirms, however, that "[t]he right to know the case to be met is not absolute": at para. 57. The question is whether there are adequate safeguards. The court reflected on this in the context of national security and foreign intelligence, at para. 61:

Fundamental justice requires substantial compliance with the venerated principle that a person whose liberty is in jeopardy must be given an opportunity to know the case to meet, and an opportunity to meet the case. Yet the imperative of the protection of society may preclude this. Information may be obtained from other countries or from informers on condition that it not be disclosed. Or it may simply be so critical that it cannot be disclosed without risking public security. This is a reality of our modern world. If s. 7 is to be satisfied, either the person must be given the necessary information, or a substantial substitute for that information must be found.

(Emphasis added)

[213] After Charkaoui, the Government of Canada implemented a process involving special advocates, similar to the mechanism used in the United Kingdom -- something alluded to by the Supreme Court of Canada as an adequate substitute. Under this system, a Federal Court judge appoints a special advocate who is not a party to the proceedings or counsel to the named person, but whose role is to protect the interests of the named person. The special advocate has access to the secret materials and may cross-examine witnesses and make submissions to the court, but may not disclose the information to the named person.

[214] We conclude, therefore, that there is no categorical exclusionary rule regarding resort to intelligence-based evidence in surrender cases. The Minister must be satisfied, however, that adequate protections exist in the requesting state to ensure the named person is subject to a fair prosecution. [page225]

[215] Here, the Minister was satisfied that was the case, and we cannot say his decision in this regard was unreasonable.

[216] The Minister began by addressing the concerns put forward by the appellant about the French system. The appellant submitted to the Minister that

the French juges d'instruction cannot properly determine the reliability of the intelligence allegations made against him because they are not allowed to know the sources of the allegations but, rather, presume that all information received from intelligence officers is reliable;

the defence cannot effectively probe or question the underlying material in an intelligence report, and intelligence officers are not required to answer when cross-examined; and

French courts use intelligence as evidence to prosecute terrorists and have admitted torture-derived statements as evidence in the past.

[217] In response to these concerns, the Minister made inquiries of the French officials. France's Ministry of Justice provided him with information regarding the features and protections of the French system afforded to the appellant. The Ministry represented that

the juge d'instruction is tasked with gathering the evidence and combines everything, including the intelligence reports, into the case dossier, of which the appellant will have a copy;

intelligence allegations alone cannot be used to refer the appellant's case to trial or to convict him, but, rather, must be corroborated by traditional evidence;

the appellant may raise concerns with the intelligence evidence in his first appearance; he may request to have the juge d'instruction question its sources during the judicial examination stage; and he may call the sources at trial;

the appellant has the right to appeal the juge d'instruction's and the trial court's decisions, including any decision not to call or challenge the intelligence sources, and the right, as well, to appeal to the European Court of Human Rights; and [page226]

the appellant will be presumed innocent until he is convicted after trial; he will have access to legal representation throughout; he can provide the juge d'instruction with any document to be included in the case dossier; he can direct the juge d'instruction to investigate on his behalf; and he can call on any witnesses he chooses.

[218] Notwithstanding the appellant's evidence challenging the adequacy of these protections and assurances, the Minister concluded that it was not his role to resolve any conflicts in how France's laws and procedures are interpreted and implemented. Rather, absent evidence of bad faith or improper motive by France, the Minister was entitled and obliged to rely on France's explanation of its legal system and its legal safeguards. The Minister was satisfied that the French system, while different from the Canadian or English common law systems, has substantial checks and balances at all stages to protect the appellant's due process rights.

[219] This was not an unreasonable approach, and it reflects the prevailing case law. To suggest that the Minister cannot rely on representations by French authorities about French law -- in the absence of compelling evidence of bad faith to the contrary -- would be to discredit a respected extradition partner with a long history of upholding the rule of law. It would be "a reflection of the gravest possible kind, not only upon the motive and actions of the responsible Government, but also impliedly upon the judicial authorities of a . . . friendly Power": Arton (Re), [1896] 1 Q.B. 108 (Div. Ct.), at p. 115 Q.B., cited with approval in Schmidt, at pp. 526-27 S.C.R.

[220] In addition, the appellant will be afforded the protection of the Convention for the Protection of Human Rights and Fundamental Freedoms, November 4, 1950, 213 U.N.T.S. 221 ("European Convention on Human Rights"), the International Covenant on Civil and Political Rights and the Convention Against Torture, which ensure the right to a fair trial and the prohibition against torture-derived evidence. France is party to all three conventions.

[221] We would not give effect to this ground of review.

(3) Dealing with the potential use of torture-derived evidence in the surrender context

(i) The appellant's position

[222] The appellant's second argument on this aspect of the judicial review is that it was unreasonable for the Minister [page227] to order his surrender to France because there is a real risk that the intelligence-sourced evidence on which France will rely in prosecuting him includes evidence obtained through means of torture.

[223] In making this submission, he and the intervenors again recognize that two competing dynamics are in play -- Canada's international comity and treaty obligations, on the one hand, and, on the other hand, the need to protect individuals against violations of their fundamental rights if that is the consequence of the surrender decision. They argue, however, that particular difficulties arise in terrorism cases where a person facing surrender alleges that evidence gathered through the efforts of international intelligence agencies has been derived through means of torture. Torture is difficult to prove, and intelligence-sourced evidence is, as discussed above, "unknown, unsourced, and uncircumstanced". It is, therefore, difficult to attack, and requiring a person facing surrender to show a sufficient link between the impugned evidence and torture to any significant standard of "proof" is unfair and unrealistic.

[224] For these reasons, the appellant and the intervenors urge us to adopt a two-stage framework for these purposes. They submit that, in the initial stage, a person facing surrender need only establish a "plausible connection" between the challenged evidence and the use of torture. Once this is done, they contend, the Minister is then obliged to undertake a further investigation and to satisfy himself or herself through appropriate inquiries of and/or appropriate assurances from the requesting state that the evidence to be relied upon at trial has not been obtained through the use of torture.

(ii) A two-step inquiry and the rationale for it

[225] We agree that a two-step approach should be adopted in cases involving the Minister's decision regarding surrender where the person facing surrender alleges that he or she may face prosecution on the basis of intelligence-sourced evidence tainted by torture. Canadian and international jurisprudence supports this view.

[226] A two-part inquiry where the use of torture-derived evidence is in play has been adopted by the Federal Court of Canada in two decisions involving the same individual: Mahjoub v. Canada (Minister of Citizenship and Immigration), [2006] F.C.J. No. 1862, 2006 FC 1503, [2007] 4 F.C.R. 247, at pp. 263-66 F.C.R. (per Tremblay-Lamer J.); and Mahjoub (Re), [2010] F.C.J. No. 900, 2010 FC 787, at paras. 51-59 (per Blanchard J.). It is also the approach favoured in the United Kingdom and in the [page228] jurisprudence of the European Court of Human Rights: see, for example, A & Others v. Secretary of State for the Home Department (No. 2), [2005] UKHL 71, [2006] 2 A.C. 221 (H.L.); and Othman (Abu Qatada) v. United Kingdom, Application No. 8139/09 (January 17, 2012) (Eur. Ct. H.R.).

[227] These authorities all agree that the initial threshold to be met by the person raising the concern about the use of torture-derived evidence is a low one. They describe this threshold in various ways as "a plausible connection", "a plausible explanation" or "a plausible reason".

[228] The same authorities are not consistent, however, in their approach to the second stage of the inquiry. Some say the onus is on the decision-maker to be satisfied on a "balance of probabilities" that the evidence to be used against the person raising the concern is not tainted by torture. Others use the terminology of "real risk" or "substantial risk". Still others refer to "reasonable grounds". Some combine all of these notions at the second stage of the inquiry.

[229] We would simplify these somewhat differing approaches by adopting a "plausible connection" for the first step of the inquiry and a "real risk" for the second step. Accordingly, in our view, the two-step inquiry is properly framed in this way: (i) the person facing surrender, and challenging such evidence, must first show that there is a "plausible connection" between that evidence and the use of torture; and (ii) if that threshold is met, the Minister is then called upon to make further inquiries and satisfy himself or herself -- on the basis of the record, any further information obtained and/or assurances received from the requesting state -- that there is "no real risk" that torture-derived evidence will be used in the proposed foreign proceeding. If the Minister is satisfied on that basis, the surrender order should be made; if not, surrender should generally be refused.

[230] Our reasons for adopting this test are as follows.

[231] First, we accept that a person facing surrender in these circumstances is placed in an untenable position if required to prove on a balance of probabilities or any similar standard that torture-derived evidence sourced through intelligence agencies will be used in the foreign proceedings. To impose a threshold of that nature in such circumstances would be unrealistic and unfair. As Lord Bingham cogently put it in A & Others, at para. 59, "[i]t is inconsistent with the most rudimentary notions of fairness to blindfold a man and then impose a standard which only the sighted could hope to meet".

[232] Second, we accept that extradition proceedings are not intended or expected to be determined on the same footing as [page229] criminal proceedings in the requesting state, for which extradition is sought. The jurisprudence is clear that "extradition proceedings are not trials", but are "intended to be expeditious procedures to determine whether a trial should be held": McVey, at p. 551 S.C.R. This view is as relevant to the surrender stage as to the extradition hearing, in our opinion. It tells against applying a trial-like threshold of proof, such as a balance of probabilities, to the surrender process.

[233] Finally -- given these factors -- the challenge is to formulate a process that honours Canada's obligations internationally and the need to defend society from the ravages of terrorism, while at the same time protecting Canadians from unacceptable violations of their Charter rights. We think the two-step process referred to above best reconciles these competing tensions.

[234] Although the test for refusing surrender on s. 7 grounds is a "strict one" and is only met in "very exceptional" cases, it is beyond debate that torture-derived evidence may not be used in legal proceedings and cannot be relied upon by a state seeking extradition or being asked to extradite. Article 15 of the Convention Against Torture -- to which both Canada and France are signatories -- makes this clear. It provides that parties are obliged to "ensure that any statement which is established to have been made as a result of torture shall not be invoked as evidence in any proceedings, except against a person accused of torture as evidence that the statement was made". Section 269.1(4) of the Criminal Code, R.S.C. 1985, c. C-46 incorporates this prohibition into Canadian domestic law. To similar effect is s. 83(1.1) of the [Immigration and Refugee Protection Act](https://www.canlii.org/en/ca/laws/stat/sc-200