R v. A.K., 2026 ONSC 2262

CITATION: R v. A.K., 2026 ONSC 2262

COURT FILE NO.: CR-25-00000674-00BR

DATE: 2026 04 17

ONTARIO

SUPERIOR COURT OF JUSTICE

B E T W E E N:

HIS MAJESTY THE KING Respondent

Carson Coughlin, for the Crown respondent

-and-

A.K. Applicant

Boris Bytensky and Adam James O’Brodovich, for the applicant

HEARD: January 12, 2026 by Video Conference

POSTSCRIPT TO REASONS ON S. 520 BAIL REVIEW APPLICATION

D.E. HARRIS

1A criminal accused comes first into contact with the judicial system when brought to court for a bail hearing. The significance of bail to our system of justice is incontrovertible. “[F]air, timely, focused and effective bail hearings …are a priority.”¹ These reasons concern an important procedural aspect of bail.

2The basic tenets and specific provisions of our modern bail system date back to the Bail Reform Act, S.C. 1970-71-72, c. 37 and are now over 50 years old. Our approach to bail and its essential structure has remained fundamentally unchanged over that period of time. Although the topic of bail has been very much in the public eye over the last several years, the issue to be examined here is not part of that controversy. No one can question that the bail system must be accurate. It is critical that accused who should be released, are released and those accused who should be detained, are detained. The system must balance public protection and safety with safeguarding the Charter of Rights and Freedoms individual right in s. 11(e) not to be denied reasonable bail without just cause.

3Ultimately, the theme of efficiency and efficacy underlies the resolution of the issue raised here. It is vitally important that bail hearings, intended to be expeditious, not get bogged down in technicalities and minutiae that may unnecessarily encumber and restrict proceedings: Toronto Star Newspapers Ltd. v. Canada, 2010 SCC 21, [2010] 1 S.C.R. 721, at paras. 23-24, 27.

4I previously detained the accused in this case. In my view, the strong and very serious prosecutorial case against him on the home invasion robbery charge demonstrated that he was a dangerous person and that surety supervision was not a sufficient comfort to reduce the risk he poses to the public: R v. A.K., 2025 ONSC 5537. The home invasion allegations involve the firing of a handgun and pistol whipping the victim in his bedroom while threatening his life.

5A statement made by the Justice of the Peace who detained the applicant at the first instance was worrisome. He said he could not impose a stricter bail than the one offered by the accused based on the holdings in two Superior Court rulings in what has been cryptically styled as “bail set not met” cases. Those cases are R. v. C.S., 2023 ONSC 6406 and R. v. Parkes, 2024 ONSC 3368. Because of its importance to the administration of criminal justice, I invited further submissions on whether those cases and the Justice of the Peace’s interpretation of them were correct.

6Parenthetically, it should be said at the outset that the phrase “bail set not met” is not a felicitous one and has little to recommend it. The phrase serves to identify a particular type of bail order, but the language is neither descriptive nor helpful. What is meant is that at a bail hearing, a Justice of the Peace or a judge, as the case may be, determines that the accused is releasable but the plan and/or the surety proposed are not sufficient. The release order is made subject to the subsequent presentation and approval of an appropriate surety in a separate proceeding. Only after approval of the surety, is the accused released. The process is bifurcated in this way. First the release order; then the approval of the surety or sureties.

7Both Crown and defence have now helpfully advanced further arguments on the “bail set not met” question. Because I have already detained the accused and held that he is not releasable no matter the plan of supervision put forward, it is clear that a stricter bail as mentioned by the justice would not alter that result. It would have made no difference in this case.

8Nonetheless, it is important that the issue be carefully examined. There have been concerns voiced for some time now that in routine matters, the “bail set not met” decisions are resulting in the clogging up of bail courts and the slowing down of usual bail procedures. The problem is significant. Some accused that ought to be released are not being released, potentially contributing to the ever growing crisis of jail overcrowding. There is a reasonable basis for concluding that the main case, C.S., is per incuriam and that a significant body of case law going back many years was not taken into account. In the circumstances, in my view, the issue merits a second look: Borowski v. Canada (Attorney General), [1989] 1 S.C.R. 342; Taylor v. Newfoundland and Labrador, 2026 SCC 5, 510 D.L.R. (4th) 195, at para. 44.

R. v. C.S., 2023 ONSC 6406

9The applicant argues that C.S. was wrongly decided and should not be followed. The issue here and in C.S. itself is said to arise only in reverse onus bail hearings in which the accused must show cause why he or she should be released from custody: see Code, s. 515(6). In C.S., a reverse onus situation, the Crown successfully applied under s. 521 of the Code to review the bail release order made by the Justice of the Peace. The defence conceded that the bail release should be vacated although they did not concede the bail set not met issue: C.S., at para. 3. Abrams J. held that a “bail set not met” order was not known to law and was made without legal authority.

10What had happened in front of the Justice of the Peace in the court below was this. No surety was available or proposed for the accused C.S. at the bail hearing. Instead, the Justice of the Peace put a release order in place to permit the accused in the future, if she was able to find someone to act as surety, to satisfy the order and be released. This was a “bail set not met” order. But what happened after was completely unsatisfactory. A few days after the bail hearing, a man presented himself to sign as surety. The approval process was ex parte and conducted by a Justice of the Peace. Only very cursory questions were asked of the surety. Very little, if any, background checks or examination of this man were conducted. The questions elicited that he did not have a criminal record, knew the accused because years ago he had given her a place to stay, advised she would listen to him when she was not under the influence of alcohol and that she would stay with him in Ottawa. She was released on bail with this man as the surety.

11It is not difficult to imagine the abuses this summary procedure could be susceptible to but there is no need to do so. The facts of C.S. itself are sufficient illustration. The surety was 50 years old and the accused was a troubled 20-year-old woman. Two years before, he had begun “dating” her despite her young age. In two previous incidents, she had alleged that he had sexually assaulted her. On another occasion, he was charged with assaulting her, including punching her in the face five times and dislocating her shoulder. He was placed on a one year peace bond in November of 2021 and required to stay away from her. There was also some evidence that he was trafficking her for a sexual purpose.

12Clearly, something went frightfully wrong when C.S. was released to a person that had likely abused her in the past. Sadly, the surety-accused relationship may have provided a legal structure permitting him to mistreat and control her by threatening to rescind the bail pledge he had signed. Needless to say, the “pull of bail” and a surety’s duty to supervise the accused could not be properly exercised in these circumstances: Canada (Attorney General) v. Horvath, 2009 ONCA 732, 248 C.C.C. (3d) 1, at paras. 40-43.

13Without at this point going through the accompanying reasoning, the judge in C.S. held that what the justice had done by fashioning a release order without approving a surety was invalid and not available as a matter of law. The judge held, at para. 34, that “[i]n summary, “bail set not met”, in a reverse onus situation, is not an order known to law. The Criminal Code does not support it. The Order is quashed” (emphasis added).

DOES HORIZONTAL STARE DECISIS PROHIBIT A SECOND LOOK AT THE ISSUE?

14I have come to the conclusion that with respect, the C.S. decision was reached per incuriam. It ought not to be followed. Furthermore, in so far as Parkes found that C.S. should not be overturned and still ought to be considered a valid precedent, with respect, I cannot agree.

15The question of whether a decision is per incuriam is governed by R. v. Sullivan, 2022 SCC 19, [2022] 1 S.C.R. 460, where Kaiser J. reiterated that the law was as stated in Re Hansard Spruce Mills (1954), 4 D.L.R. 590 (B.C.S.C.), at p. 592. There are only three possible justifications for departing from a decision of a judge sitting at the same level:

(i) the rationale of an earlier decision has been undermined by subsequent appellate decisions;

(ii) the earlier decision was reached per incuriam (”through carelessness” or “by inadvertence”); or

(iii) the earlier decision was not fully considered, e.g., taken in exigent circumstances.

Sullivan, at para. 75.

16Applying the Sullivan tests to C.S., it was not as though one or two anecdotal cases were missed and not considered. It was an entire line of well-established authority which was bypassed. The one case relied on to arrive at the conclusion that the bail orders in question are invalid does not actually stand for that proposition. If anything, this case is an example of a bifurcated proceeding and establishes its legitimacy.

17To those in the legal profession of a certain vintage, the conclusion in C.S. is startling. In the 1990s and going back many years, sureties were often approved in a proceeding separate from the bail hearing itself. This was standard procedure. While not as common now, the practice is still prevalent. It does not seem that the judge in C.S. was informed of this reality.

18I hasten to add that this does not mean that what happened in C.S. was not troubling. Clearly, something went badly awry in that instance. But the adage that bad facts make bad law comes to mind when the execution of a bail order marred by serious errors in the surety approval process leads to a blanket holding, despite an unbroken line of jurisprudence to the contrary, that a separate surety approval process is never permissible and is legally invalid.

19The first and most important case confirming the validity of a release order issued without naming or approving a surety is that of S.C. Hill J. in R. v. Brooks, (2001), 153 C.C.C. (3d) 533 (Ont. S.C.), decided 25 years ago. It was an accused’s bail review under s. 520 of the Code following his detention by a Justice of the Peace. It was a Crown onus hearing. The Justice of the Peace refused to accept surety questionnaires that had been filed out and required the sureties to appear in court. But the sureties were unable to attend the hearing.

20Hill J. found that the Justice of the Peace had erred. His reasons recognize a judge’s right to issue a bail order without naming a surety. It is worthwhile quoting them at length:

34 A prospective surety may attend court and testify at a s. 515 show cause hearing. The surety is then subject to cross-examination. Such a witness, or the accused person, or another witness may testify as to knowledge of another person or persons who are suitable and prepared to stand as surety should the court order release. Not infrequently, a prospective surety takes time off work to be interviewed at the office of counsel for the accused ahead of the show cause hearing. He or she may not be in a position to lose another day from the workplace sitting and waiting to be called to testify in a bail court with a dozen to two dozen contested hearings, or worse still, finding that the relevant hearing is adjourned because it cannot be reached on the congested court list.

35 There is no legal requirement that a surety be present in court and testify in order for judicial interim release to be considered. This is especially so where the Crown bears the onus of establishing why continued detention is justified. By virtue of S.C. 1985, c.19, s. 84(2), the Code was amended with the addition of what is now s. 515(2.1):

Power of justice to name sureties in order — Where, pursuant to subsection (2) or any other provision of this Act, a justice, judge or court orders that an accused be released on his entering into a recognizance with sureties, the justice, judge or court may, in the order, name particular persons as sureties.

From the 1972 Bail Reform Act to the time of this amendment, the court ordering release would set out in its order the number of sureties and the amount by which they were bound. In the pre-amendment era, the justice supervising release of the accused, once the accused was in a position to comply with the order, would interview the prospective surety or sureties and, if suitable, proceed with execution of the recognizance. This is a function long exercised by a Justice of the Peace.

36 The inclusion of s. 515(2.1) confers a permissive authority in the court — “...may, in the order, name particular persons as sureties”. The court is not required to do so and, because a prospective surety is not in court or is as yet unidentified, the court is often not in a position to in effect pre-approve a surety to be named in the order. It then falls to the justice exercising the compliance jurisdiction contemplated by s. 519 of the Code to approve, or not, the sureties before release can occur.

37 The observations here should not be misunderstood to suggest that, on the facts of a particular case, a judicial officer could not justifiably, and within the exercise of her or his judicial duties, wish to see a prospective surety called to testify. A commonplace example is where the accused proposes to live with his surety. However, a blanket requirement that all prospective sureties, in every case, must appear before the court at the show cause hearing amounts to an abuse of discretion.

38 In this case, the prospective sureties were working and unable to attend court. The applicant and his mother were able to identify who the proposed sureties were. The surety questionnaires contained an abundance of information about the individuals. The presiding justice ought to have at least reviewed the documents in an effort to determine whether they were properly receivable as credible and trustworthy evidence pursuant to the authority of s. 518(1)(e) of the Code. Regardless of whether the questionnaires proved sufficient to name the relevant persons as sureties in the order pursuant to s. 515(2.1), the court would in any event have a sense of the existence of support for the applicant’s release. In the present case, the court’s reference to counsel for the applicant having surety statements “not...filed with the court” seems to indicate an adverse inference of sorts. Counsel tried to file the questionnaires as exhibits and the court declined even to receive them for review as to whether they could properly be marked as exhibits. [Emphasis added.]

21As Hill J. points out in the second part of para. 35, separating the bail release hearing from the process for the approval of the surety, is not unusual. The separate surety approval process “is a function long exercised by a JP.” This process is governed by s. 519 of the Code: Brooks, at para. 36; R. v. Shaefer, [1999] O.J. No. 2175, at paras. 14, 42.

22Hill J. went on to expand on his Brooks reasons in other cases: see R. v. Villota, (2002), 163 C.C.C. (3d) 507 (Ont. S.C.); R. v. Wynter, 2015 ONSC 2426. In Villota, at para. 81, Hill J. wrote:

As canvassed in the Brooks decision, the presence of a prospective surety in court is not a jurisdictional or essential prerequisite to the conduct of a judicial interim release hearing. While the defence case may be tactically disadvantaged by failing to call a prospective surety or sureties to testify, in many cases the surety approval process is best left to an informal interview by the releasing Justice of the Peace. There is abundant authority for this approach: R. v. Gillespie, supra at 142-5; R. v. Schaefer (1999), 97 O.T.C. 161 (S.C.J.) at para. 29-31, 34, 36, 43; Annotation: Bail in Criminal Cases, E. Armour (1927), 47 C.C.C. 1 at 7, 12-12, 16; The Law of Bail in Canada, supra at 294-5. Restricting consideration of the sufficiency of sureties to the bail hearing stage inevitably lengthens bail hearings and compounds the congestion already existing in busy courts: Fundamental Justice and the Approval of Sureties by the Crown, G.T. Trotter (1987-88), 30 C.L.Q. 238 at 247, 249.

23Following these cases, in his important judgment in R. v. Tunney, 2018 ONSC 961, Di Luca J. was highly critical of inefficiencies in the system as well as the overuse of sureties. On the question of surety approval, he both recognized and approved of the bifurcated bail hearing practice, writing:

40 In many locations, Justices of the Peace are available to approve sureties outside of court either by attendance “over the counter” or in chambers. Using out of court surety approval processes contributes to various efficiencies within the system. In particular, it makes bail hearings simpler, shorter and more focussed. It makes it easier and more convenient for sureties to attend court in order to secure the release of accused persons. It spares precious court time, which in the era of Jordan is at a premium.

41 In some jurisdictions, the surety approval process has historically taken place in court as part of the bail hearing. At times, this process works appropriately, particularly in serious cases where an exceptional surety may make the difference between detention and release. At other times, this approach has not only caused unnecessary delays and inconvenience, but it has also resulted in unfair practices. Indeed, it appears that there have been times when the surety approval process has taken on a needlessly adversarial tone, which has created the impression that the proposed surety is somehow on trial: see Bail and Remand in Ontario, at pp. 32-34; Berger & Stribopoulos, at pp. 317-318. To state the obvious, bail hearings are not trials. Procedures and practices that unduly discourage potential sureties from coming forward to offer their services are to be avoided: see Canada (Minister of Justice) v. Mirza, at paras. 47-48. Uniform or regular use of in-court surety approval is such a practice.

42 More significantly, the use of the in-court surety approval process in ordinary cases creates a risk that the bail hearing will morph into a constructive reverse onus scenario. It also creates a risk that the issue of releasability will be confused with the issue of surety suitability. Over 15 years ago, Duncan J. addressed this very issue in R. v. Cole, [2002] O.J. No. 4662 (Ont. C.J.). At paras. 18-20, he explained:

The determination of surety suitability can be done either by the justice at the bail hearing itself or, if an order for release on a recognizance with sureties is made at the hearing, by a justice before whom the recognizance is entered into: See Trotter, supra, p. 200; R. v. Brooks, supra. The practice in this jurisdiction has been (and apparently still is) to have sureties attend the bail hearing and testify before the justice. Justice Hill has been trying to direct the bail courts in this jurisdiction that, contrary to their ingrained practices, sureties need not be present and testify before the Justice making the release/detention decision. His decision in Brooks, supra, apparently went largely unheeded, necessitating his comprehensive decision in R. v. Villota, (2002) 2002 49650 (ON SC), 163 C.C.C. (3d) 507. He identified a number of serious problems that arise from the practice, including the routine adjournment of bail hearings because sureties are not present, and delays and backlog caused by bail hearings being prolonged while sureties are examined and cross-examined.

24Many cases have followed Brooks and the other cases to recognize the validity of bifurcated bail proceedings including R. v. A.K., 2020 ONSC 1555, at para. 21; R v Renaud, 2010 ONSC 5300; R. v. Schultz and R v A.F., 2020 ONSC 2880, at para. 37. No cases disapproved of or deviated from these authorities until C.S. came along.

25The case law overwhelmingly and without exception acknowledges and affirms the practice of bifurcated hearings. Trotter J.A., one of the leading authorities on bail, also confirms their validity and historical legitimacy in his textbook, The Law of Bail in Canada, 3rd ed. (Toronto: Thomson Reuters). In the chapter discussing sureties (§ 7:11), he says:

[T]he process for determining suitability [of sureties] may involve viva voce evidence, according to the standard examination-in-chief/cross-examination format. But it need not be addressed in this formal manner. Other methods that may be used. Section 518(1)(a) permits the justice or judge to make inquiries about the accused “on oath or otherwise,” just as s. 518(1)(e) allows a decision based on “evidence considered credible and trustworthy.” The justice may also take into consideration relevant matters “agreed on by the prosecutor and the accused or his counsel”: s. 518(1)(d). Moreover, the newly enacted surety declaration in s. 515.1 should also prove to be helpful tool.

26Trotter J.A. excerpted the key passages from Brooks and Villota quoted above with approval. He also approved of a judicial comment made in R. v. Smith-Lowe, 2022 ABQB 494, at paras. 33-34 that examining sureties in court can be invaluable and enhance public confidence in the administration of justice. He concluded (§ 7:11):

Whether sureties should be examined during a bail hearing is a matter that should be decided on a case-by-case basis. In some cases, where the allegations are not serious, but it is determined that a surety release is appropriate, it may not matter who will stand as the accused’s surety(ies). In other cases, it will be crucial. Justices and judges should determine whether it is necessary for the issue to be addressed in court and then consider how it may be effectively done. Some of the less formal procedures referred to above, including surety declarations, should be considered.

If particular sureties are not named by the justice or judge conducting the bail hearing, then it falls to the Justice of the Peace who takes the recognizance to determine this issue. This usually occurs when the proffered surety attends before a justice with a view to seeking the release of the accused. At that time, the justice will make inquiries, perhaps under oath, to determine whether the surety is suitable or not. If the justice concludes that the proffered surety is unsatisfactory, the accused remains in custody. [Footnotes omitted]

27The judgment in C.S. did not refer to any of the cases on point nor to Trotter J.A.’s text. The passages referred to above were in Trotter J.A.’s book at the time C.S. was released and had been there for many years. Nor was there advertence to the decades long practice of sureties being approved in a proceeding separate from the bail hearing.

28The C.S. decision was written after the accused’s bail order had already been revoked, with the consent of the defence, and she had been detained: C.S., at paras. 2-4. It appears that the issue was one the Crown (not Mr. Coughlin) felt was important and decided to pursue. A lengthy factum was written focussing only on the bail set not met issue. The Crown in their factum, a copy of which was provided to me, argued that “[t]he bail set not met process and the ex parte approval of sureties is not recognized in law. This court has the authority, and the Crown submits an obligation, to address the issue.”

29The Crown factum argued forcefully that ex parte approval of sureties is never appropriate either in reverse onus or Crown onus situations. The troubling facts of C.S. were used to illustrate the dangers. In their factum, there was some brief reference to Wynder, Villota and Brooks, but there was no real discussion of those cases. At no place in their factum was the continued validity of bifurcated proceedings, as unequivocally established in the case law and commentary, brought to the attention of Abrams J. That is unfortunate.

30It is not totally clear what input, if any, came from the defence on the bail set not met issue. But based on the C.S. reasons, it does not appear that the defence joined the issue. It is not surprising, given the Crown’s approach in their factum and the lack of a lis on the issue, that the leading authorities were not cited or discussed in Abrams J.’s decision. The practice of separate surety approval was not mentioned and seems to have entirely escaped notice. There was reference in the reasons to the bail review decision in R. v. Mallaley, 2020 ONSC 7178, which had been cited and discussed in the Crown’s factum.²

31Abrams J., in C.S., reviewed Mallaley saying that no surety was proposed at the original bail hearing and so a detention order was made. At the bail review stage, according to the court’s summary at para. 33, a surety was proposed. The court, according to Abrams J., “accepted that the J.P. found the accused was ‘releasable’ but not on the plan proposed and thus, properly issued a detention order.” Abrams J. held, as the Crown had argued in their factum, that Mallaley was “a simple example of the permissible and proper approach to be followed in reverse onus hearings.”

32With respect, the discussion in C.S. seriously misconstrues Mallaley. In Mallaley, it was conceded by the Crown on the bail review that the Justice of the Peace at the original hearing had concluded that the accused was releasable. However, there being no surety proposed, the plan was insufficient and the accused was detained. Things had not changed by the time of the bail review except that the accused now said he could find a surety. It was held on the bail review in Mallaley that if a surety was produced, they could be approved in a future proceeding:

40 While I contemplated releasing the Applicant on his own recognizance (see R v. Ismail, 2020 ONSC 5519 (Ont. S.C.J.)), the seriousness of the allegations and his criminal record counsel against this. Supervision is certainly preferable.

41 Mr. Quayat [for the Crown] argued that the best way to proceed would be to adjourn this proceeding and reconvene a hearing with the proposed surety. He would then have the opportunity to cross-examine her.

42 I have decided not to take that route. Justices of the Peace have extensive experience with the surety approval process: see R. v. Tunney, 2018 ONSC 961 (Ont. S.C.J.) at paras. 39-42. I am confident that in accord with these reasons, a Justice of the Peace will ensure that any proposed surety, given the important role they are to play, will be carefully vetted. [Emphasis added.]

33Properly understood, in Mallaley, a release order was made but the surety approval was to be effectuated independently. In retrospect, given the seriousness of the allegations, the Crown may have been correct that cross-examination of the proposed surety before the bail review judge ought to have been the course followed.³ But the point is that Mallaley demonstrates the use and the validity of a separate surety approval by a justice after a bail release order has been issued. Ironically, what occurred in Mallaley is, if anything, a concrete illustration of a so called “bail set not met” order being valid and legitimate.

34In summary, the conclusion is inescapable that C.S. was reached by inadvertence: see Re: Ramos, 2025 ONCA 820, at paras. 36-39. The C.S. decision conforms well with the definition in Sullivan, at para. 77, of what constitutes a per incuriam decision,

[A] judge can depart from a decision where it was reached without considering a relevant statute or binding authority. In other words, the decision was made per incuriam, or by inadvertence, a circumstance generally understood to be “rare” (see, e.g., The Owners, Strata Plan BCS 4006 v. Jameson House Ventures Ltd., 2017 BCSC 1988 (B.C. S.C.), 4 B.C.L.R. (6th) 370, at para. 132). The standard to find a decision per incuriam is well-known: the court failed to consider some authority such that, had it done so, it would have come to a different decision because the inadvertence is shown to have struck at the essence of the decision. It cannot merely be an instance in which an authority was not mentioned in the reasons; it must be shown that the missing authority affected the judgment (Rowe and Katz, at p. 19). [Emphasis added.]

R. v. PARKES, 2024 ONSC 3368

35The subsequent case of Parkes was also a Crown s. 521 bail review in a reverse onus situation. While not itself per incuriam, in my view, with respect, Barrett J. erred in not overruling C.S. and holding that it was per incuriam.

36In Parkes, the Crown argued that the court was bound to conclude as the judge in C.S. had that a bail “set not met” is a not a valid order. The Justice of the Peace in the original bail hearing was satisfied that the accused was releasable but did not believe the release plan was adequate. A release order was fashioned specifying that the surety could be approved at a later date. Several days later, on the record in open court, and after Crown cross-examination, a surety was approved and the accused was released. The Justice of the Peace issued written reasons in which he refused to follow the C.S. case.

37Barrett J., on the bail review in Parkes, agreed with the Crown and held that the Justice of the Peace, as a matter of vertical stare decisis, was bound to follow the decision in C.S. This holding is not challenged.

38As an aside, it is interesting to note that although the Barrett J. in Parkes found that the “bail set not met” order should be vacated as it was invalid, the Crown conceded and Barrett J. agreed that the surety who had been originally approved was adequate to assuage any concerns on the secondary grounds. Mr. Parkes was released on precisely the same terms and with the same surety as he had been originally released with under the bail set not met order: see Parkes, at paras. 38-40. Nothing changed.

39In C.S. as well, the Crown in their factum on the bail review conceded that the accused was releasable on an appropriate plan being put forward. The record does not indicate whether that occurred. It is clear that both C.S. and Parkes are not so much about substantive requirements for bail release but instead are concerned with the procedural steps that should be taken in naming and approving sureties.

40The ultimate conclusion reached in Parkes appears to be that C.S. was per incuriam. However, Barrett J. found that the result in the case would have been the same even if the authorities recognizing and affirming the bifurcated practice had been considered. Therefore, Barrett J. concluded that C.S. ought not to be overturned and should be followed. Barrett J. reasoned, at para. 37:

In this case, the issues of concern identified by the Justice of the Peace suggest that Abrams J.’s decision [in C.S.] was reached per incuriam as the decisions of Wynter and Mallaley were either not mentioned or not fully considered. However, as explained in Sullivan at paras. 77 and 79, it is insufficient to simply point to a failure by the jurist to consider a relevant authority. The circumstances in which a decision is found to have been made per incuriam are “rare”. The question is whether the failure to consider a prior decision impacted the “essence of the decision” such that the outcome would be different. Here too, I am satisfied that there is no basis to depart from the doctrine of horizontal stare decisis. Abrams J.’s finding that a BSNM order is invalid in the circumstances of a reverse onus bail was grounded in statutory interpretation; specifically, the interplay between ss. 515(2), 515(2.1) and515(6). The decisions of Wynter and Mallaley are not based on a statutory analysis. Accordingly, I am satisfied that the high threshold test for demonstrating that Abrams J.’s decision would have been different had he fully addressed each of these decisions is not met. [Emphasis added.]

41With respect, I cannot agree. If the decisions in the cases starting with Brooks had been considered in C.S., it is virtually inevitable that the bifurcated process of bail described in those cases would have been found to be valid. It would have been recognized that issuing a bail release order and approving sureties can be accomplished in separate proceedings.

42Contrary to what was said in Parkes, the declaration that bail set not met orders are invalid and unknown to law, was the “essence of the decision” as that phrase is used in Sullivan. The essence was certainly not a question of whether the ultimate result would have been different: i.e. that the Crown’s bail review would not have been allowed. Any judge would have disapproved of the process that took place in C.S. and the inadequate questioning of the surety by the justice. Amongst other inadequacies, there were clear questions about suitability raised by the surety’s answers which ought to have been followed up on.

43The defence quite rightly conceded the inadequacies in the surety approval process and agreed that the bail order should be rescinded. The Crown and Abrams J. then took up the issue of whether a bail set not met order was valid across the board. The result of the bail review itself was never in question.

44If Abrams J. in C.S. had known of the long standing practice of approving sureties in a separate proceeding and was directed specifically to the judicial cases and commentary confirming it and extolling its importance to the process of an efficient bail system, his decision on that question would undoubtably have been different. The inadvertence committed in C.S. therefore “struck at the essence of the decision.”

45I also cannot agree, with respect, that the means by which the decision was reached, whether by statutory interpretation or by some other means, is of any relevance to the question of the soundness of the C.S. decision. That is an extraneous consideration.

46The departure from precedent and practice in C.S. and its consequence to the administration of justice are matters of serious concern. Here, the inadvertence cancelled out a whole line of jurisprudence extending back many years. There is every reason for overturning a decision which stands contrary to prior judicial authorities and runs against a common practice.

THE REASONS IN C.S. TO HOLD THAT BIFURCATED PROCEEDINGS ARE NOT PERMITTED WERE UNCONVINCING

47The process of issuing a bail order but approving of sureties separately warrants another look. I am not bound by C.S., as it is per incuriam.

48The first stage in establishing whether C.S. was wrongly decided is an examination of the authorities beginning with Brooks and going on to the other cases and Trotter J.A.’s commentary. Those authorities confirm the historical legitimacy of this mode of proceeding. That has already been done above and need not be repeated. These cases and practices are more than adequate support for bifurcated proceedings in appropriate cases.

49But furthermore, looking at the reasoning in C.S. and the holding that this way of proceeding is unauthorized, demonstrates that, with respect, it was seriously flawed. I agree with Mr. Bytensky that it seems to be based virtually exclusively on the “ladder principle” not applying to reverse onus hearings. Although I agree that it does not apply, I fail to understand how it is a necessary correlative of that conclusion that it is improper to make a release order and approve the sureties only later.

50The ladder principle is codified in s. 515(2.01) of the Code. That section reads:

Imposition of least onerous form of release

(2.01) The justice shall not make an order containing the conditions referred to in one of the paragraphs (2)(b) to (e) unless the prosecution shows cause why an order containing the conditions referred to in the preceding paragraphs for any less onerous form of release would be inadequate.

51Similarly, s. 515(1) in combination with s. 2 establishes a presumption that an accused should be released without conditions unless the Crown can show otherwise. In addition, Abrams J. relied on the permissive power to name sureties in s. 515(2.1). That provision reads:

Power of justice — sureties

(2.1) If, under subsection (2) or any other provision of this Act, a judge, justice or court makes a release order with a requirement for sureties, the judge, justice or court may name particular persons as sureties.

52It was a combination of these two provisions which led to the core holding in C.S., at paras. 29-32:

Reliance on s. 515(2.1), which appears to be the basis for the “bail set not met” order, is misplaced…

Section 515(2.1) is inextricably linked to ss. 515(2) and (2.01). Section 515(2)(c) provides for the assignment of sureties. Section 515(2.01) in turn codifies the “ladder principle”, mandating that the imposition of sureties only be imposed if the court has considered the imposition of lesser conditions in s. 515(2). Where the court is so satisfied, it may name a surety pursuant to s. 515(2.1).

However, the “ladder principle” - as codified in s. 515(2.01) - does not apply in a reverse onus hearing. Thus, it follows that s. 515(2.1), linked inextricably to ss. 515(2) and (2.01), does not apply in a reverse onus bail hearing: R. v. Sakhiyar, 2018 ONSC 5767, at paras. 4-7; R. v. Ishmael, 2019 ONSC 596, at paras. 27-34.

Put differently, to the extent that s. 515(2.1) is the basis for “bail set not met”, it is inapplicable in a reverse onus hearing. In R. v. T.T., [2018] O.J. No. 6985 (S.C.), at para. 29, the court noted that in a reverse onus hearing it is not the court that “devises” a bail plan. Rather, one is proposed and accepted or rejected: “I note that this was a reverse onus bail hearing, which placed the burden on Mr. T.T. to convince the justice that the risk could be managed in the community. It is not up to the court to devise a bail plan. The reverse onus bail hearings raise very different considerations”.

53With respect, I fail to see the logic of this reasoning. Beginning with the ladder principle, the Supreme Court’s judgment in R. v. Antic, 2017 SCC 27, expounded on its importance. The now Chief Justice, Wagner J. wrote, at paras. 29-30:

The Bail Reform Act also codified what is now known as the “ladder principle”. This Act set out possible forms of release, which were ordered from the least to the most onerous. The ladder principle generally requires that a justice not order a more onerous form of release unless the Crown shows why a less onerous form is inappropriate. In other words, the ladder principle means “that release is favoured at the earliest reasonable opportunity and... on the least onerous grounds”: R. v. Anoussis, 2008 QCCQ 8100, 242 C.C.C. (3d) 113(C.Q.), at para. 23, per Healy J.C.Q. (as he then was).

The ladder principle and the authorized forms of release remain a central part of the Canadian law of bail and are now enumerated in s. 515(1) to (3) of the Code.

54Later, in summarizing the pertinent principles, Wagner J. said that the ladder principle must be “adhered to strictly… each rung of the ladder must be considered individually and must be rejected before moving to a more restrictive form of release”: Antic, at paras. 67(d), (f).

55There is a good deal of judicial thought considering whether the ladder principle applies to reverse onus bail hearings. The cases go both ways but the weight of authority is that the ladder principle is inapplicable in the reverse onus context: see Trotter, § 6:11, The Ladder Principle, at n. 11. Despite there being little reasoning in the case law to support this conclusion, I believe it is persuasive.

56In a Crown onus situation, each restriction on liberty is another rung going up the ladder and must be justified by the Crown. Similarly, the presumption established by ss. 515(1) and (2) that release on a no condition bail is the default position is premised on the same principle as the ladder in s. 515(2.01). The lowest restriction on bail release is presumed.

57But in a reverse onus, the defence must demonstrate that the accused should be released. The logic of the ladder principle in s. 2.01 does not hold. The ladder section refers to “the prosecution show[ing] cause why an order containing the conditions referred to in the preceding paragraphs for any less onerous form of release would be inadequate” (emphasis added). The ladder principle is at bottom a principle that bail conditions ought to be the least restrictive imposition upon liberty available unless the prosecution can prove otherwise. But in a reverse onus, the onus is borne by the accused, not by the Crown.

58The plain language and meaning of the ladder principle provision read in full context demonstrate that it cannot be adapted to fit reverse onus situations: R. v. Wilson, 2025 SCC 32, 507 D.L.R. (4th) 573, at paras. 32-36. This appears to be confirmed in R. v. Zora, 2020 SCC 14, [2020] 2 S.C.R. 3, at para. 101, where it was said, “[o]ther than in reverse onus situations, the Crown should understand, and if asked, be able to explain why proposed bail conditions are necessary, reasonable, least onerous, and sufficiently linked to the risks in s. 515(10)” (emphasis added).

59Finally, the language of the preamble to the reverse onus provision in s. 515(6) demonstrates the difference between a Crown onus and a reverse onus situation:

Order of detention

(6) Unless the accused, having been given a reasonable opportunity to do so, shows cause why the accused’s detention in custody is not justified, the justice shall order, despite any provision of this section, that the accused be detained in custody until the accused is dealt with according to law, if the accused is charged …

60One is the inverse of the other. The Crown, if it is their onus, must justify going up the ladder; the defence, if it is their onus, must justify going down the ladder.⁴

61But the inapplicability of the ladder principle in reverse onus hearings is of little moment and, in any case, is tempered by several other considerations. Like with a Crown onus bail, a judge should never impose conditions more restrictive than necessary. Pre-trial detention diminishes the presumption of innocence. Before being found guilty by a judge or jury, the accused without bail has their liberty curtailed: R. c. Pearson, [1992] 3 S.C.R. 665, pp. 683-89; R. v. Hall, 2002 SCC 64, [2002] 3 S.C.R. 309, at para. 47, Iacobucci J. dissenting; adopted by majority in Toronto Star Newspapers Ltd., at para. 51. That is a concern that weighs in the balance at every bail hearing. The least restrictive release necessary to fulfill bail objectives ought to be imposed in accord with the protection afforded by s. 11(e) of the Charter not to be denied reasonable bail without just case: R. v. Kilbride, 2022 ONSC 1074, at para. 23.

62The basic precepts of bail are immutable wherever the onus falls. For example, s. 515(2.03) reads:

For greater certainty, before making an order requiring that the accused have a surety, the justice shall be satisfied that this requirement is the least onerous form of release possible for the accused in the circumstances.

63Furthermore, in a similar vein, s. 493.1 of the Code requires that the primary consideration on bail ought to be towards release. This applies to both Crown and reverse onus cases.⁵ The provision reads:

PRINCIPLE OF RESTRAINT

493.1 In making a decision under this Part, a peace officer, justice or judge shall give primary consideration to the release of the accused at the earliest reasonable opportunity and on the least onerous conditions that are appropriate in the circumstances, including conditions that are reasonably practicable for the accused to comply with, while taking into account the grounds referred to in … 515(10), as the case may be.

64While dating only from 2019, this provision is a codification of the common law which has advocated restraint in the imposition of bail for many years. The Supreme Court has stated that “in Canadian law, the release of accused persons is the cardinal rule and detention, the exception”: R. v. St-Cloud, 2015 SCC 27, [2015] 2 S.C.R. 328, at para. 70: see R v. Al-Adhami, 2020 ONSC 6421, at para. 57; R. v. E.B., 2020 ONSC 4383, at para. 23. This comment was not restricted only to Crown onus bails. The Supreme Court emphasized in Zora, at para. 26, that restraint in bail matters is a constitutional value:

Many interveners drew attention to the widespread problems which continue to exist, even after this Court’s decision in Antic, with the ongoing imposition of bail conditions which are unnecessary, unreasonable, unduly restrictive, too numerous, or which effectively set the accused up to fail. Any such practice offends the principle of restraint which has always been at the core of the law governing the setting of bail conditions. Restraint has a constitutional dimension, a legislative footing, and is not only recognized in case law, but was also recently expressly reinforced by the amendments that came into force on December 18, 2019. Section 493.1 now explicitly sets out a “principle of restraint” for any interim release decisions … [Emphasis added.]

65“The reversing of the onus does not change the presumption of release”: R. v. Peddle, [2022] N.J. No. 9 (Nfld. L. Prov. Ct.), at para. 10. The Supreme Court has enshrined the paramountcy of release within the constitutional protection cast by s. 11(e) of the Charter. It was held in Zora, at para. 20:

From a constitutional perspective, most bail conditions restrict the liberty of persons who are presumed innocent and impose a risk of further criminal liability on those persons because of the failure to comply offence under s. 145(3). Therefore, the setting of bail conditions must be consistent with the presumption of innocence and the right not to be denied reasonable bail without just cause under s. 11(e) of the Charter (see Antic, at para. 67; R. v. Tunney, 2018 ONSC 961, 44 C.R. (7th) 221 (Ont. S.C.J.), at para. 36). Section 11(e) protects both the right not to be denied bail without “just cause” and the right to bail on reasonable terms and conditions (R. c. Pearson, [1992] 3 S.C.R. 665 (S.C.C.), at p. 689; R. c. Morales, [1992] 3 S.C.R. 711 (S.C.C.), at p. 735; Antic, at paras. 36-41). The s. 11(e) right is “an essential element of an enlightened criminal justice system” that “entrenches the effect of the presumption of innocence at the pre-trial stage of the criminal trial process and safeguards the liberty of accused persons” (Antic, at para. 1). The presumption of innocence is “a hallowed principle lying at the very heart of criminal law. ... [that] confirms our faith in humankind” (Antic, at paras. 66-67(a), quoting R. v. Oakes, [1986] 1 S.C.R. 103 (S.C.C.), at pp. 119-20; see also R. v. Myers, 2019 SCC 18 (S.C.C.), at para. 1). The presumption of innocence is only satisfied in the bail process when the requirements of s. 11(e) are met (Pearson, at pp. 688-89; Morales, at p. 748). As described by Andrew Ashworth and Lucia Zedner, the presumption of innocence and the protection of liberty rights mean that “the state should presume each person to be harmless ... therefore it is in principle wrong to take coercive measures against people for preventive reasons unless there are very strong justifications for doing so” (Preventive Justice (2014), at p. 53). The Charter therefore protects accused persons from unreasonable terms and conditions of bail. [Footnote omitted.]

66In addition, it was suggested in C.S., para. 23, relying on R. v. T.T., [2018] O.J. No. 6985 (S.C.), at para. 29, that in reverse onuses, the justice ought not to “devise the plan.” The bail suggested by the accused--its quantum, sureties, and conditions--can only be accepted or rejected by the justice. There is no ability to amend what has been recommended by the accused. This notion is part of the erroneous approach that postulates that a reverse onus is a totally different animal than is a Crown onus bail. That is a fundamental misconception. The onuses are different, that is true and important. But there is no justification for imposing a rule in reverse onuses different than the law and practice in Crown onus bails. In the bail context as in many other contexts, a judge or Justice of the Peace has the power to alter counsel’s recommendations and do justice in a particular case if he or she feels it is appropriate. Liberty is at stake. Technicalities and rules without articulable underlying rationales have no place at a bail hearing. Artificial impediments cannot be justified. Bail procedures must remain flexible and “procedural straightjackets that would interfere with the administration of justice” should be avoided: St-Cloud, para. 138.

67Flexibility is particularly necessary in the ever evolving, fluid question of bail release for the reasons given by Doherty J.A. in R. v. Jaser, 2020 ONCA 606, 152 O.R. (3d) 673, at paras. 52-53:

Circumstances relevant to bail are dynamic and can change significantly as a case progresses and time passes: see R. v. Zora, 2020 SCC 14at para. 92. Bail orders are inherently more interim in nature than final. The integrity of the criminal justice process would suffer if the finality principle was afforded the same weight in bail reviews as it is given on appeals from convictions.

St. Cloud adopts a flexible, more receptive approach to fresh evidence on bail reviews under s. 520 and s. 521. That approach reflects the inherently interim nature of bail decisions and accommodates Charter principles underlying the presumption of innocence and access to reasonable bail.

68Returning to the heart of the issue, I do not understand how the ladder principle, whether or not it applies in reverse onus cases, or the discretionary power to name sureties, has anything to do ultimately with whether there can be a bail order made but the sureties approved in a subsequent proceeding. Reverse onuses differ from Crown onuses and alter a basic predicate of a bail hearing. But that is a substantive difference, not a procedural one. There is no coherent reason why a bail hearing can be separated from surety approval in a Crown onus situation but not in a reverse onus situation. The only case making that distinction is C.S.

CONCLUSION

69In my view, a bail release order can be issued without a surety being approved as part of that process. A so-called “bail set not met” order is valid. There are several aspects however that require care and caution.

70In a serious case where a surety is of significant importance as a component of a bail release plan, the most rigorous method is for the judicial officer to scrutinize the proposed surety in person and for them to be cross-examined by the Crown.

71The judgment in C.S., as part of its conclusion that sureties must appear in court during the bail hearing and must be named, relied heavily on s. 11(i) of the Crown Attorneys Act, R.S.O. 1990, c. C.49. That provision provides that when a person is in custody and charged with a criminal offence, the Crown Attorney “shall … inquire into the facts and circumstances, and satisfy himself or herself as to the sufficiency of the surety or sureties offered, and examine and approve of the bail bonds where bail is ordered” (emphasis added). This provision applies regardless of whether the onus is on the Crown or the accused. The court in C.S. went on to emphasize the importance of cross-examination to test evidence in our system of justice: C.S., at paras. 54-58. But no special rules for reverse onus cases can be derived from the legislative framework or the common law.

72In serious cases where release relies to a substantial degree on the integrity and supervisory capacity of a surety, cross-examination will generally be called for. But there are some cases in which it will not be necessary. As Trotter J.A. says in the quote at para. 26 above, this issue must be evaluated on a case-by-case basis. For routine matters, the Crown may well agree that there is no need for the surety to be cross-examined as long as the justice performs the approval process with reasonable care. The decision in C.S. currently prohibits proceeding in this manner even when there is consent on both sides.

73In other situations where the Crown is of the view that the surety should be fully scrutinized, whether cross-examination should be conducted ought to be up to the presiding justice at the bail hearing. If the justice is of the view that there should be cross-examination, it may be that the bail hearing can be adjourned to a fixed date to examine the surety.

74Surety declarations are required by the Criminal Code: see s. 515.1. If the surety is unavailable for the bail hearing, a surety declaration could fill in for their absence and convey sufficient information to name the surety. Alternatively, declarations will assist a justice in a separate approval process.

75For these reasons, in my view, C.S. should no longer be followed as it is per incuriam and was wrongly decided.

Footnotes

1. 1“Ontario Court of Justice Practice Direction Regarding Bail Hearings” issued under Rule 5.1 of the Criminal Rules of he Ontario Court of Justice, effective date, June 1, 2026
2. 2I was the judge who wrote the Mallaley decision.
3. 3Court records show that Mallaley never found a surety and was not released on bail.
4. 4Bill C-14, An Act to amend the Criminal Code, the Youth Criminal Justice Act and the National Defence Act (bail and sentencing), 1st Sess, 45th Parl, 2025-2026, currently being studied at committee in the Senate, would clarify in s. 23(1) that the ladder principle in s. 515 (2.01) does not apply to reverse onus bail hearings.
5. 5Bill C-14 in s. 14 would make the principle of restraint inapplicable in reverse onus situations.