Court of Appeal for Ontario

CITATION: R. v. Singh, 2025 ONCA 843

DATE: 2025-12-05

DOCKET: COA-23-CR-0203 & COA-23-CR-0408

Tulloch C.J.O, Roberts and Monahan JJ.A.

DOCKET: COA-23-CR-0203

BETWEEN

His Majesty the King

Appellant

and

Sukhvir Singh

Respondent

DOCKET: COA-23-CR-0408

AND BETWEEN

His Majesty the King

Appellant

and

Simranjeet Narang

Respondent

Diana Lumba and Ryan McSheffrey, for the appellant

Glen Henderson and Jag Virk, for the respondent, Sukhvir Singh

No one appearing for the respondent, Simranjeet Narang

Heard: May 6, 2025

On appeal from the stays entered by Justice Nyron Dwyer of the Ontario Court of Justice on February 1, 2023, with reasons reported at 2023 ONCJ 123, 523 C.R.R. (2d) 84 (COA-23-CR-0203), and on March 10, 2023 (COA-23-CR-0408).

Tulloch C.J.O.:

A. INTRODUCTION

[1] Section 11(b) of the Charter guarantees the right to be tried within a reasonable time. Courts must reconcile this fundamental right with two of the administration of justice’s most compelling imperatives: trying co-accused jointly and conducting large project prosecutions. The flexible framework in R. v. Jordan, 2016 SCC 27, [2016] 1 S.C.R. 631, achieves this very goal. It recognizes that joint trials are ordinarily in the interests of justice and that complexity is sometimes unavoidable in project prosecutions, while holding the Crown to its duty to prevent and mitigate delay. Applying Jordan mechanically without meaningfully accounting for these interests risks fragmenting cases into a multiplicity of proceedings, compounding systemic delay, and undermining the vital role of project prosecutions. Properly applied, Jordan prevents such outcomes while still safeguarding timely justice for the accused.

[2] This project was massive. Both respondents, along with dozens of others in multiple jurisdictions, were charged with drug offences following an 11-month interagency investigation of cross-border drug trafficking. The Crown tried them together with two other co-accused who dealt with the same high-level international drug trafficker. The consequent voluminous disclosure and scheduling challenges pushed the case modestly above the 18-month ceiling.

[3] The application judge declined to find any exceptional circumstances and stayed the charges. He concluded that the Crown should have tried the respondents individually and that the charges were straightforward, not complex.

[4] I would allow the Crown’s appeals and set aside the stays because the joint trial and the project’s scale justified the additional time taken. In my respectful view, the application judge did not meaningfully account for these vital interests. His preferred alternative to a joint trial, severance, was not a panacea. Rather, it would have undermined the very interests joint trials safeguard – advancing the search for truth, strengthening public confidence in the courts, and combatting systemic delay. By focusing narrowly on the relative simplicity of the individual charges, he overlooked the structural complexity of the project as a whole. The complexity of the case caused everything to take longer – even for the respondents – and easily justifies the remaining modest delay.

B. BACKGROUND

[5] In May 2020, York Regional Police launched “Project Cheetah,” a major investigation into cross-border drug trafficking. Working closely with the Royal Canadian Mounted Police, Peel Regional Police, and United States federal authorities, officers conducted surveillance, obtained judicial authorizations, and identified three high-level American drug traffickers and their Canadian contacts, including the respondents, Simranjeet Narang and Sukhvir Singh. On takedown day, in April 2021, police arrested both of them along with 27 others in Ontario, three in British Columbia, and the three high-level traffickers south of the border.

[6] The Crown decided to proceed jointly. It grouped the respondents and two other accused, Lakhpreet Brar and Balwinder Singh Dhaliwal, together as the “Ravi Group.” The connections linking them together were strong – they were charged as co-conspirators, shared a common high-level supplier, participated in multiple observed transactions involving the same undercover officer, and faced similar trafficking charges. Mr. Narang, for instance, participated in transactions with not only Mr. Singh, but also Mr. Brar and Mr. Dhaliwal.

[7] The investigation’s sweeping scale generated massive disclosure. Over the course of 14 months, the Crown produced over 86,000 files – roughly 300 gigabytes of material – including undercover communications, surveillance, search warrants, and forensic reports. The Crown’s initial mega productions required seven months to complete, and defence review time and follow-up requests added a further seven months.

[8] The joint trial and enormous quantities of disclosure required additional time. The challenge of coordinating multiple defence counsels’ schedules caused the loss of several pre-trial conference and trial dates, extending the time to trial by 107 days. As well, the hundreds of hours required to review the immense productions left defence counsel unable to elect a mode of trial at the first pre-trial conference, requiring a 35-day adjournment.

[9] Trial in the Ontario Court of Justice was ultimately set for March 2023 – 116 days over the presumptive ceiling for Mr. Narang and 175 days above it for Mr. Singh, after accounting for an uncontested deduction and the timing of charges. The respondents and their co-accused applied for a stay.

[10] The application judge stayed the proceedings. He refused to deduct the 107 days of joint trial scheduling challenges. Despite acknowledging that the Crown’s choice to proceed jointly was logical, he suggested that it should have pursued individual trials or split the Ravi Group into sub-groups to prevent delay. Citing the “time consuming and difficult” task of reviewing the voluminous disclosure to make an informed election, he also declined to deduct the 35-day adjournment as defence delay – a conclusion which I would uphold despite the Crown’s challenge to it.[^1] But he pivoted at the case complexity stage, asserting that the “main point” was not the project’s overarching complexity, but instead the charges’ simplicity.

[11] The Crown appealed. Since it could not locate the two co-accused and Mr. Narang did not respond, this court allowed it to proceed with the appeals against the two respondents and heard them despite Mr. Narang’s non-participation.

C. ANALYSIS

[12] The Crown asks this court to set aside the stays of proceedings and remit the matters for trial, arguing that the application judge erred in his assessment of exceptional circumstances. It submits, first, that the application judge should have deducted the 107 days of joint-trial scheduling delay as a discrete exceptional circumstance. Second, it contends that the application judge failed to recognize that the complexity of the case justified the remaining delay in excess of the Jordan ceiling.

[13] I would allow the appeal. I agree with the Crown on both issues. Although findings regarding exceptional circumstances generally attract deference, a correctness standard applies here because the application judge misapplied the governing legal principles: R. v. Zahor, 2022 ONCA 449, at para. 79; Ontario (Labour) v. Nugent, 2019 ONCA 999, 384 C.C.C. (3d) 189, at para. 28, leave to appeal refused, [2020] S.C.C.A. No. 53.

(1) The Joint Trial Delay Should Be Deducted

[14] On the first issue, I accept the Crown’s position that the application judge should have treated the 107 days arising from joint-trial scheduling difficulties as a discrete exceptional circumstance and deducted them accordingly. I begin by reviewing the applicable legal principles before turning to their application on the record.

(a) Jordan Accommodates Joint Trials

[15] Courts have consistently emphasized that joint trials are strongly preferred over separate trials because they promote the proper administration of justice. Trying co-accused together – particularly in conspiracy cases or where the charges arise from a common series of events – advances the truth-seeking function in a way that separate trials may undermine. Joint trials also safeguard public confidence by reducing the burdens multiple proceedings impose on witnesses, jurors, and the community, and by avoiding the risk of inconsistent verdicts that may erode trust in the justice system. In addition, they achieve significant systemic efficiencies by preventing duplicative proceedings that would increase costs, strain judicial resources, and exacerbate delay. Accordingly, joint trials are the presumptive rule and severance the exception: R. v. Crawford, 1995 138 (SCC), [1995] 1 S.C.R. 858, at paras. 30-32; R. v. Chow, 2005 SCC 24, [2005] 1 S.C.R. 384, at paras. 47-48; R. v. Last, 2009 SCC 45, [2009] 3 S.C.R. 146, at paras. 16-18; R. v. Sarrazin (2005), 2005 11388 (ON CA), 75 O.R. (3d) 485 (C.A.), at para. 59; R. v. Rai, 2019 BCCA 377, 381 C.C.C. (3d) 1, at para. 154.

[16] Jordan accommodates these compelling policy considerations. The Supreme Court has recognized that joint trials may require additional time and may, in appropriate cases, justify delay in excess of the ceiling. Delay resulting from a joint trial that serves the interests of justice should, therefore, be deducted as a discrete exceptional circumstance where it was unforeseen or reasonably unavoidable and where the Crown acted reasonably to mitigate it: Jordan, at para. 77; R. v. Vassell, 2016 SCC 26, [2016] 1 S.C.R. 625, at para. 6; R. v. Tran, 2023 ONCA 532, 429 C.C.C. (3d) 55, at para. 40.

[17] At the same time, Jordan reconciles this strong policy preference with an accused’s right to be tried within a reasonable time. The “interests of justice” analysis accommodates both considerations, and the Crown remains obliged to take reasonable steps to move the case forward and to mitigate joint-trial delay. In some circumstances, this may require severance – particularly where a joint trial unduly complicates the proceeding, causes substantial delay, or effectively holds an accused who wishes to proceed promptly hostage to the delays of co-accused: Jordan, at paras. 77, 79; R. v. Manasseri, 2016 ONCA 703, 429 C.C.C. (3d) 55, at paras. 323, 367-75, leave to appeal refused, [2016] S.C.C.A. No. 513; R. v. Gopie, 2017 ONCA 728, 140 O.R. (3d) 171, at paras. 138-42, 171.

[18] However, the Crown is not required to forgo the systemic benefits of joint trials by severing proceedings as a matter of routine whenever a case approaches the ceiling. Severance is not invariably a solution, particularly where its benefit to the accused’s interest in a timely trial is marginal. Overuse of severance risks fragmenting complex project cases and other multi-accused proceedings into numerous separate trials, thereby compounding demands on already strained courts. This would undermine the truth-seeking function, erode public confidence, frustrate Jordan’s objective of systemic reform, and perpetuate the very systemic delay that Jordan sought to address: Rai, at paras. 153-54.

[19] To avoid this systemic spiral, courts must give meaningful effect to the strong policy favouring joint trials. That policy will frequently justify modest scheduling delays – an inherent feature of joint trials that Jordan expressly contemplates: Tran, at paras. 43-44; Gopie, at paras. 166, 169-70, 174.

(b) This Joint Trial Must Be Accommodated

[20] The application judge ought to have deducted the 107 days attributable to joint-trial scheduling challenges, as this period constituted a discrete exceptional circumstance.

[21] Intervention is warranted because the application judge was required to address the threshold inquiry: whether proceeding by way of a joint trial was in the interests of justice. Instead, he proceeded on the basis that joint trials involve trade-offs, without determining whether the strong presumption in favour of a joint trial was displaced.

[22] A joint trial was presumptively appropriate and clearly in the interests of justice. The respondents and their co-accused were alleged co-conspirators, connected through a common supplier, multiple related transactions, and extensive overlapping disclosure. Proceeding jointly served the compelling public interests in truth-seeking, fostering public confidence, and promoting systemic efficiency.

[23] The remaining preconditions for a discrete exceptional circumstance were also met. The 107-day delay was caused by the unavailability of various defence counsel for judicial pre-trials and trial dates. The Crown could not reasonably have avoided or mitigated these scheduling complications, which, as the application judge himself recognized, are an inherent feature of many joint trials: R. v. Jurkus, 2018 ONCA 489, 363 C.C.C. (3d) 246, at para. 67, leave to appeal refused, [2018] S.C.C.A. No. 325.

[24] The application judge’s suggestion that the Crown should have pursued separate trials, either individually or by dividing the “Ravi Group” into sub-groups, overlooks that severance is not a universal remedy. Here, severance would have undermined the truth-seeking function and heightened the risk of inconsistent verdicts. The proposal to try Mr. Singh and Mr. Brar separately, for example, would have excluded Mr. Narang, who allegedly interacted with both, as well as Mr. Dhaliwal, Mr. Singh’s associate. Severance would also have compounded systemic delay: it would have required witnesses to testify multiple times, involved additional judges, and consumed substantially more court resources. Such an approach would run counter to Jordan’s systemic objectives, not advance them.

[25] The application judge appears to have overlooked these systemic considerations by focusing primarily on the respondents’ s. 11(b) interests. That was an error. The analysis requires reconciling the strong policy favouring joint trials with the accused’s right to a timely trial. In this case, the balance favoured the former. The 107-day extension reflected ordinary scheduling realities of a joint trial, and unlike Manasseri, neither respondent was held hostage by a co-accused’s dilatory conduct. As Tran confirms, modest additional delay of this nature, a fraction of the delay at issue in Manasseri, must be accommodated and does not displace the presumption in favour of a joint trial: Tran, at paras. 43-44.

[26] Deducting the 107-day period substantially narrows the above-ceiling excess. For Mr. Narang, it reduces the overage from 116 days to only nine days. For Mr. Singh, the overage drops from 175 days to just 68 days.

(2) Case Complexity Justifies the Remaining Delay

[27] The next issue is whether the complexity of this case justifies the remaining delay above the ceiling – nine days for Mr. Singh and 68 days for Mr. Narang. In my view, it does. I begin by outlining the governing legal principles before turning to their application to the record.

(a) Case Complexity Can Accommodate Project Prosecutions

[28] To establish case complexity, the Crown must demonstrate two elements. First, it must show that the nature of the evidence or issues required an inordinate amount of preparation or trial time sufficient to justify the remaining delay above the ceiling. This threshold demands particular complexity – something beyond the baseline contemplated by Jordan’s ceilings. Once particular complexity is established, the degree of complexity required to justify delay is proportionate to the quantum of net delay: shorter net delays require less complexity, while longer delays require more. Second, the Crown must show that it implemented a concrete, reasonable plan to minimize delay: Jordan, at paras. 77-79; R. v. Cody, 2017 SCC 31, [2017] 1 S.C.R. 659, at paras. 63-64; R. v. Chan, 2019 ABCA 82, 82 Alta. L.R. (6th) 1, at para. 24; R. v. Picard, 2017 ONCA 692, 137 O.R. (3d) 401, at paras. 64-69, leave to appeal refused, [2018] S.C.C.A. No. 135.

[29] Complexity is frequently invoked where the state advances a project prosecution involving expansive investigations, voluminous disclosure, and numerous accused persons. As Karakatsanis J.A. (as she then was) observed, such prosecutions “play an important role in the administration of justice” and serve as a “necessary response to the increasing sophistication of organized criminal activity”: R. v. Khan, 2011 ONCA 173, 270 C.C.C. (3d) 1, at paras. 30, 82, leave to appeal refused, [2011] S.C.C.A. No. 195.

[30] Jordan’s accommodation of case complexity reflects the systemic importance of project prosecutions. The doctrine aims to make room for time-intensive proceedings – including organized crime cases – that cannot reasonably be completed within the presumptive ceilings. Although the focus is on complexity rather than offence gravity, large multi-accused organized crime trials are among the cases most likely to satisfy this threshold: Jordan, at paras. 77-78, 81; Nugent, at paras. 31-33; R. v. Pearce, 2021 NSCA 37, 405 C.C.C. (3d) 75, at para. 145.

[31] Jordan also directs courts to assess complexity holistically over the full life cycle of the case. Complexity should not be viewed in isolation at the trial stage. Early-stage complexity – arising from the breadth of disclosure and number of accused in large projects – may justify longer delay even for “minor players” whose individual trials appear more streamlined by the end: R. v. Lopez-Restrepo, 2018 ONCA 887, 369 C.C.C. (3d) 56, at paras. 37-40.

[32] As a result, courts must approach case complexity in a manner that is both purposive and realistic in the context of project prosecutions. As de Sa J. has explained, the sheer scale of such cases means they inevitably take longer to organize and advance. It would be unfair, and unrealistic, to expect them to proceed at the pace of ordinary prosecutions: “Understanding R. v. Jordan: A New Era for s. 11(b)” (2018) 66 Crim. L.Q. 93, at p. 97. The fact that a Crown plan anticipates exceeding the ceiling does not itself render the plan unreasonable: Nugent, at para. 39.

[33] At the same time, complexity cannot operate as a shield for Crown complacency. Jordan requires the Crown to move project cases forward as expeditiously as reasonably possible. Complexity is never presumed, even in large projects; the Crown must prove it on the record and must take reasonable steps to reduce delay. Not every prosecution within a project will be complex, and some may be completed within the ceiling even if others require more time. Moreover, complexity cannot justify unreasonable Crown decisions, ineffective planning, or choices that themselves generate delay: Jordan, at para. 79; Picard, at paras. 64-69.

(b) This Project Justified the Additional Time

[34] The complexity of this prosecution justified the modest net delay that remained above the ceiling. Unfortunately, this was overlooked by the application judge, resulting in a legal error.

[35] The application judge erred by failing to situate the respondents’ charges within the broader procedural context. He treated the case as uncomplicated because the allegations against these two accused were relatively focused. But that is not determinative. They were prosecuted as part of a large, multi-accused mega-project involving extensive investigative steps and enormous disclosure. Those structural features can render a proceeding complex even where individual accused – as here – face comparatively straightforward charges. His further suggestion that the lack of novel legal issues precluded complexity misstated the test: complexity arises from the nature and volume of the evidence, the scale of the investigation, and the organizational demands of a project prosecution – not solely from difficult questions of law.

[36] Viewed holistically, this was a particularly complex case. Project Cheetah was a cross-border, multi-agency investigation involving wide-ranging surveillance, wiretap authorizations, and dozens of warrants. The resulting disclosure was vast: over 300 gigabytes, which took more than a year to assemble and produce. As the application judge elsewhere recognized, these features created unavoidable delays because disclosure management was “time consuming and difficult” for both sides. This is precisely the paradigm of mega-project complexity contemplated in Jordan.

[37] The caution in Jordan and Cody that voluminous disclosure alone may be insufficient does not alter the analysis. Here, the volume of disclosure was intertwined with the project’s cross-border scope, the number of investigative techniques used, and the dozens of defendants implicated. This project was substantially larger than those in Cody or Jordan – the 86,000-plus files far exceeded the roughly 20,000 pages in Cody, and the number of implicated individuals was more than triple that in Jordan. Moreover, there was no evidence that this prosecution or others within the project could reasonably have proceeded within the ceiling.

[38] This overarching complexity readily justifies the relatively modest above-the-ceiling delay. While the respondents’ charges were not technically difficult, in the circumstances of this case that factor may only limit the complexity justification for longer periods of delay; it does not negate the need to accommodate the additional time warranted by a mega-project of this scale at its early stages.

[39] The record also demonstrates that the Crown implemented a concrete, reasonable plan to minimize delay. The application judge accepted that the Crown took the standard steps required in complex projects: assigning appropriate resources, devoting substantial time to disclosure management, organizing the many accused into tailored prosecution groups, and using case management tools to streamline the pre-trial and trial processes: R. v. Bulhosen, 2019 ONCA 600, 377 C.C.C. (3d) 309, at para. 82, leave to appeal refused, [2019] S.C.C.A. No. 370 (Kompon), and [2019] S.C.C.A. No. 423 (Bulhosen).

[40] The application judge nevertheless faulted the Crown because disclosure remained time-consuming. That reasoning was incorrect. The standard is one of reasonableness, not perfect efficiency, and the Crown is not required to complete a complex project case within the ceiling or eliminate all delay: Nugent, at paras. 39, 44-45. Having found no unreasonableness in the Crown’s plan, the very factors the judge identified – the time required to prepare extensive disclosure and the defence’s ensuing requests for further disclosure and review – reinforce rather than undermine the conclusion that this was a complex prosecution.

D. DISPOSITION

[41] In the result, I would allow the appeal, set aside the stays of proceeding, and remit the matters to the Ontario Court of Justice for trial.

Released: December 5, 2025 “M.T.”

“M. Tulloch C.J.O.”

“I agree. Roberts J.A.”

“I agree. P.J. Monahan J.A.”

[^1]: I reject the Crown’s argument that the application judge applied an overly stringent test. Despite once suggesting that “substantial” disclosure was needed, read fairly his reasons show that he applied the correct test – that disclosure must be sufficient to allow an informed election: R. v. Girimonte (1997), 1997 1866 (ON CA), 37 O.R. (3d) 617 (C.A.), at p. 625, leave to appeal granted but appeal discontinued, [1997] S.C.C.A. No. 606.