COURT OF APPEAL FOR ONTARIO

CITATION: R. v. Aubin, 2026 ONCA 324

DATE: 20260507

DOCKET: C70301

Fairburn A.C.J.O., Monahan and Osborne JJ.A.

BETWEEN

His Majesty the King

Respondent

and

Brandon Aubin

Appellant

Richard Litkowski, for the appellant

Jeremy Streeter, for the respondent

Heard: April 13, 2026

On appeal from the convictions entered by Justice Rick Leroy of the Superior Court of Justice, without a jury, on June 29, 2020, with reasons reported at 2020 ONSC 4031, and from the sentence imposed on October 21, 2021.

Monahan J.A.:

I. OVERVIEW

[1] The appellant was the only person home when a fire broke out in the basement of his mother’s residence on the morning of April 6, 2017. The Crown’s theory was that the appellant caused the fire by producing butane honey oil (BHO), a form of concentrated cannabis resin, in the basement earlier that morning. The Crown alleged that while making the BHO, the appellant had introduced highly flammable butane into the air. An unidentified ignition source had caused the butane to ignite, leading to a flash fire and the combustion of other flammable material. The Crown’s case was circumstantial and depended primarily on the evidence of Jason Williams (“Williams”), an investigator with the Office of the Fire Marshal for Ontario (the “Fire Marshal”), that the fire had been caused by the discharge of butane vapour shortly before the fire began.

[2] The trial judge accepted the Crown’s theory and convicted the appellant of two counts of arson, contrary to ss. 433(a) and 434 of the Criminal Code, R.S.C. 1985, c. C-46, and unlawfully producing cannabis resin, contrary to s. 7(1) of the Controlled Drugs and Substances Act, S.C. 1996, c. 19.

[3] The appellant’s sole ground of appeal is that he received ineffective assistance from his trial counsel. The claim is based on the fact that trial counsel failed to retain and/or call an expert at the appellant’s trial in respect of the origin of the fire. Important context to this appeal is that trial counsel came into possession of a report (the “Matte Report”) authored by Normand Matte (“Matte”), which concluded the fire was caused by a faulty connection between an extension cord and lighting for an aquarium. Matte was retained by the property insurer.

[4] Matte was qualified as an expert witness in a pre-trial voir dire, but trial counsel did not seek legal aid funding to retain and call Matte to testify at trial. The appellant argues that as a result, the trial judge had to disregard the Matte Report which undermined his defence. The appellant submits that given the scientific complexity of the case and the centrality of the Crown’s expert evidence to its theory, the defence needed its own expert to provide full answer and defence. In his view, the lack of a competing expert caused a miscarriage of justice by compromising both the appearance of a fair trial and the reliability of the verdict.

[5] In order to succeed, the appellant must meet a three-prong test that includes factual, performance, and prejudice components. The focus of this appeal is on the prejudice component, which includes consideration as to whether (i) the ineffective representation undermined the appearance of a fair trial or (ii) had counsel been competent, whether there is a reasonable probability the result would have been different. As I explain below, the appellant cannot succeed in either branch of the prejudice component. First, there was no allegation of pervasive incompetence which could give rise to trial unfairness. Second, although the trial judge did not admit the Matte Report into evidence, he considered it in detail, its substance was put to the Crown’s expert, and he concluded that the Matte Report or the theory of the origin of the fire expressed therein did not undermine the reliability of Williams’ conclusion that the cause of the fire was the introduction of butane into the basement environment. The trial judge therefore concluded that even if Matte had testified, the result would have been the same. I come to the same conclusion. I would therefore dismiss the appeal.

II. BACKGROUND

[6] As noted above, the Crown’s case depended upon Williams’ evidence that the fire was caused by the introduction of butane vapours into the basement. Williams based this conclusion on the fact that the fire pattern evidence indicated that a flash fire had taken place. He testified that a flash fire only occurs when the fuel is in vapour form. After eliminating the possibility of other gases such as natural gas or methane, and after considering the physical evidence found in the area (including a BHO extractor containing cannabis, numerous empty butane cylinders, and a receptable dish containing trace amounts of BHO), Williams concluded that the cause of the fire was the ignition of butane vapours in the basement. He also found that the butane must have been introduced into the air shortly before the fire began, or else it would have dispersed to below flammable range.

[7] The appellant testified at his trial. He admitted to consuming BHO in the past, and that he previously attempted to produce BHO, but that he was unsuccessful. He denied ever attempting to produce BHO at the residence where the fire occurred. He did admit that he went to the basement on the morning of the fire to obtain cannabis. However, the appellant denied that he produced BHO that morning, and therefore denied that he recklessly introduced dangerous butane vapours into the basement environment.

[8] As noted above, the appellant’s trial counsel obtained a copy of the Matte Report which concluded that the fire was caused by the failure of an extension cord connection. Since Matte had been retained by the appellant’s mother’s insurer rather than by the appellant himself, Matte appeared in his personal capacity on the voir dire pursuant to a subpoena.

[9] Although the trial judge qualified Matte as an expert on the voir dire, trial counsel did not retain him and Matte did not testify at trial. The trial judge therefore ruled that the Matte Report was not properly offered as evidence and had to be disregarded. Nevertheless, the trial judge used the findings in the Matte Report to test Williams’ conclusions. In a comment clearly demonstrating his concern for the presumption of innocence, the trial judge explained that if any flaws or ambiguities became apparent in Williams’ analysis, “it would not do to summarily dismiss Mr. Matte’s opinions”.

[10] Thus, in reviewing Williams’ evidence as to the cause of the fire, the trial judge considered in detail Matte’s conclusion that the cause of the fire was a faulty extension cord connection. The trial judge identified a number of serious shortcomings in the Matte Report, including the fact that Matte was working with incomplete information, having only inspected the scene two weeks after the fire when it was compromised by fire mitigation initiatives. Moreover, Matte accepted as true the appellant’s statement that he did not access the basement on the morning of the fire and thus did not consider the possibility that the onset of the fire was related to the processing of marijuana products by the appellant.

[11] On the other hand, the trial judge was satisfied that Williams “brought considerable credible [and] reliable expertise” and “followed the evidence and considered various hypotheses and tested them”. He therefore accepted Williams’ opinion that the fire had been caused by the introduction of butane into the basement on the morning of the fire, very close in time to when the fire began.

[12] The appellant acknowledged that he had experience with BHO production in the past and admitted that he went into the basement on the morning of the fire (contrary to the facts assumed by Matte). There was no one else home at the time. The trial judge found that the appellant’s testimony was self-serving and gratuitous, and “unsupported by the facts on the ground”. He therefore rejected the appellant’s exculpatory evidence that he did not produce BHO on the morning of the fire. The trial judge found that the Crown proved beyond a reasonable doubt that the appellant recklessly introduced butane vapour into the air when producing a small amount of BHO on the morning of the fire, which ultimately caused the fire that ensued. The appellant was therefore found guilty of two arson offences and one drug production offence.

III. ISSUE ON APPEAL

[13] As noted above, the appellant’s sole ground of appeal is that he received ineffective assistance of counsel at his trial. In particular, he argues that trial counsel’s failure to retain and call Matte (or some other expert witness) fell markedly below a reasonable standard of professional judgment and competence, undermining the appellant’s defence and causing a miscarriage of justice.

[14] Sadly, trial counsel passed away in September 2022 and thus we do not have any direct explanation from him as to why he failed to retain and call an expert witness, and specifically why he did not seek legal aid funding to retain Matte, even though he had successfully sought to have Matte qualified as an expert on the voir dire. However, with the consent of the Crown, the court accepted the appellant’s fresh evidence attesting to some of the circumstances surrounding Matte and his expert report. The Crown accepts that the record establishes the following salient facts:

• Trial counsel came into possession of the Matte Report in January 2019, the month before the qualification voir dire;

• Matte was qualified on that voir dire, and the parties and the trial judge had the Matte Report;

• On March 5, 2019 (approximately one month before the beginning of the trial), Matte sent trial counsel an estimate for the cost to retain him to testify at trial of $12,200;

• Trial counsel told the appellant and his mother that legal aid would not pay for Matte to testify and that he was not going to ask them to; and

• There is no record of trial counsel having made a request to legal aid to fund either Matte or another expert in this case.

Additionally, the appellant’s mother affirmed that trial counsel told her that she and the appellant would need to pay the $12,000 fee themselves if they wanted Matte to attend the trial.

[15] The appellant argues that if his counsel had applied for legal aid funding and the request been approved, Matte or a comparable expert could have provided the court with a different explanation of the cause of the fire and provided the defence with assistance in preparing for trial, specifically the cross-examination of Williams. He submits that trial counsel’s failure to make such an application amounted to incompetent representation, undermining the appearance of trial fairness and compromising the reliability of the verdict.

IV. DISCUSSION

A. Governing Legal Principles

[16] This court has taken a cautious approach to claims based on the alleged incompetence of trial counsel. As Doherty J.A. observed in R. v. Joanisse (1995), 1995 3507 (ON CA), 85 O.A.C. 186 (C.A.), at para. 66, leave to appeal refused, [1996] S.C.C.A. No. 347, such claims can be easily made and “[i]t would be a rare case where, after conviction, some aspect of defence counsel’s performance could not be subjected to legitimate criticism”. Doherty J.A. cautioned that appeals should not devolve into “forensic autopsies of counsel’s performance at trial”, which would not only render convictions ephemeral but would also “discourage vigorous and fearless representation at trial and encourage defensive advocacy aimed more at protecting counsel from subsequent criticism than advancing the cause of his client”: Joanisse, at paras. 66-67. Given these considerations, the appellant’s burden in establishing ineffective representation is not easily discharged and reaching the threshold for appellate intervention is a “high bar”: R. v. Hartling, 2020 ONCA 243, 150 O.R. (3d) 224, at para. 74; R. v. D.M.G., 2011 ONCA 343, 105 O.R. (3d) 481, at para. 101.

[17] Against this backdrop, an appellant who advances a claim of ineffective assistance of counsel is required to establish three things:

(i) the facts on which the claim is grounded (the factual component);

(ii) the incompetence of the representation provided by trial counsel (the performance component); and

(iii) a miscarriage of justice as a result of the incompetent representation by trial counsel (the prejudice component).

See R. v. Girn, 2019 ONCA 202, 145 O.R. (3d) 420, at para. 91.

[18] The proper analytical order is generally component (i), followed by (iii), and then (ii). Once the factual component has been established, the court will first consider the prejudice component, since if this has not been satisfied, there is no reason to subject the performance of counsel to judicial inquiry: R. v. G.D.B., 2000 SCC 22, [2000] 1 S.C.R. 520, at para. 29; R. v. Fiorilli, 2021 ONCA 461, 156 O.R. (3d) 582, at para. 49. I recognize that this court has applied a modified analytical approach in cases where the primary submission on appeal is that trial fairness has been undermined by the totality of the representation: see R. v. Nnane, 2024 ONCA 609, at para. 11. As will become clear below, I do not find this is such a case, since there is no allegation of pervasive incompetence by trial counsel.

[19] The prejudice component of the test is what is materially at issue in this appeal. This prong examines the impact of counsel’s incompetence on both the appearance of trial fairness and the reliability of the result.

[20] In cases where the claim is that counsel’s performance affected trial fairness without necessarily undermining the reliability of the result, counsel’s incompetence must have been “pervasive” and amount to “conduct which permeates and infects counsel’s entire performance”: Joanisse, at para. 76. Claims under this branch must show a defect which is “‘so serious that it shakes public confidence in the administration of justice’”: R. v. White, 2022 SCC 7, [2022] 1 S.C.R. 64, at para. 9, citing R. v. Davey, 2012 SCC 75, [2012] 3 S.C.R. 828, at para. 51. The trial counsel’s conduct must have risen to a level such that it “destroys the fairness of the adjudicative process at trial, thereby amounting to a constructive denial of the assistance of counsel”: Fiorilli, at para. 55.

[21] Alternatively, where the claim is that trial counsel’s incompetence affected the reliability of the result, the claim is made out only where the appellant can show that there is a “‘reasonable probability that, but for counsel’s unprofessional errors, the result of the proceedings would have been different’”: Joanisse, at para. 79, citing Strickland v. Washington, 104 S. Ct. 2052 (1984), at p. 2068. A reasonable probability “lies somewhere between a mere possibility and a likelihood” and is established “when the reviewing court is satisfied that because of counsel’s incompetence, the verdict cannot be taken as a reliable assessment of the appellant’s culpability”: Joanisse, at para. 80; see also Fiorilli, at para. 59; R. v. S.T., 2024 ONCA 572, at para. 43. The ineffective assistance must affect a central issue at trial. There will not be a miscarriage of justice on the reliability branch unless the appellant “can inject doubt into the facts surrounding the verdict” which is achieved through “probative evidence of sufficient persuasive weight to dislodge a verdict’s foundation”: see R. v. Sararas, 2022 ONCA 58, 160 O.R. (3d) 321, at paras. 44-49.

B. Application

[22] I am satisfied that the appellant has established on a balance of probabilities the factual component of his ineffective assistance of counsel claim. It consists of counsel’s failure to seek legal aid funding to retain Matte, despite the fact that Matte had been qualified as an expert and the Matte Report reached a conclusion that apparently contradicted, at least in part, the conclusion put forward by Williams, the Crown’s expert. I say “in part” since, as discussed below, the Crown submits with some merit that the two expert conclusions were not inconsistent in any event.

[23] The next step of the analysis is to inquire whether a miscarriage of justice has occurred because either the appearance of trial fairness or the reliability of the verdict was undermined.

1. Trial Fairness Branch

[24] First, I consider the trial fairness branch. I see no basis upon which it can be concluded that the alleged incompetence of trial counsel resulted in an unfair process, either in appearance or substance. The appellant’s allegation is discrete, namely, that trial counsel failed to seek legal aid funding to retain Matte or a comparable expert. It is not alleged that trial counsel’s incompetence was pervasive or permeated his entire performance. For example, there is no suggestion that trial counsel was ineffective in cross-examining Crown witnesses, preparing for and leading defence evidence, or making submissions at trial. The appellant’s singular objection is that trial counsel did not seek legal aid funding for an expert, thereby depriving the appellant of the benefit of competing expert evidence at trial.

[25] In these circumstances, I would not find that the failure to call a reply expert “destroyed the fairness of the adjudicative process” or “amounted to a constructive denial of the assistance of counsel”, to use the language from Fiorilli. The Crown’s expert witness was ably challenged by trial counsel, including with the conclusions of the inadmissible Matte Report (as discussed above), and the appellant availed himself of his right to provide his version of events concerning the introduction of butane vapours into the air on the morning of the fire. This was not a case where the appellant advanced no defence when he had one available to him and the trial was “one-sided” as a consequence. The appellant clearly received partisan advocacy from his trial counsel throughout the proceeding. The trial judge simply did not believe the appellant’s evidence that he did not produce BHO in the basement on the morning of the fire, which is a credibility finding owed deference on appeal. The fact that the appellant’s defence was not accepted does not render the trial unfair.

[26] As noted above, given the passing of trial counsel, we did not have the benefit of any explanation from him as to why he did not try to call evidence from Matte at trial. Nor is there any evidence in the record from Legal Aid Ontario as to whether it would have authorized funding in any event.

2. Reliability Branch

[27] Turning to the reliability branch of the prejudice component, the issue is whether the appellant has demonstrated a reasonable probability that had trial counsel requested legal aid funding for an expert, the verdict would have been different. The trial judge considered this very issue and concluded that even if Matte had testified, it would not have made any difference in the result. I agree.

[28] As discussed above, far from simply disregarding the Matte Report, the trial judge considered it in detail and identified a number of serious shortcomings and caveats in Matte’s analysis, some of which were expressly recognized by Matte himself. Matte focused on how the fire was ignited, concluding that it was due to a faulty extension cord. But as Williams pointed out (and the trial judge accepted), an ignition source on its own does not cause a fire. The necessary elements of combustion are an ignition source, fuel, and oxygen. As Williams explained in his testimony, “the fire cause is the circumstances that bring those things together” (emphasis added).

[29] The trial judge illustrated this point through the following example, based on Williams’ trial evidence:

…[W]hen a person pours gasoline into a confined area and something sparks, for example, the hot water tank heater switches on to ignite the gasoline vapour, the resulting flash or even explosion was not caused by the ignition; rather by the introduction of gasoline to the contained area. No gasoline no fire. The igniting spark was just the ignition source.

[30] Because Matte focused only on identifying the ignition source of the fire, he failed to consider whether the actual cause of the fire was the introduction of butane vapour into the basement by the appellant. This despite the fact that Matte was aware that butane and equipment related to the processing of marijuana were seized following the fire. Matte understood that the introduction of such materials into the environment could have been implicated as the cause of the fire, particularly because butane is, in his words, a “highly flammable gas that can readily be ignited”. But Matte wrote in his report that “it is not possible to comment any further on this possibility” because he was “not allowed to access the evidence collected by the [Office of the Fire Marshal] as part of the official investigation”.

[31] Moreover, it is clear from Matte’s fire cause analysis that the evidence he collected from the appellant about his activities on the morning of the fire was built into the report’s conclusions. Critically, the appellant advised Matte that he did not access the basement on the morning of the fire. When Matte considered the possibility of butane vapours as the ultimate cause of the fire, he opined as follows:

Other human activities to consider in this case are related to illicit activities related to the culture, handling and processing of marijuana in the basement of the house. Based on limited information provided by representatives of the [Office of the Fire Marshal], equipment potentially related to the grow-up and processing of marijuana were seized as part of the official investigation. Given the nature of the products recovered (butane and possibly other types of equipment and supplies related to the grow-up processing of marijuana products), such activities could conceivably be related to the onset of a fire in the basement provided that such activities were being conducted in the basement on that morning. [Emphasis added.]

The significance of the appellant’s statement was clearly not lost on Matte, as he explicitly included the following caveat in his ultimate conclusion:

The circumstantial information collected by the main witness, Mr. Brandon Aubin, indicates that he did not access the basement in the morning and was sleeping at the time of the onset of the fire. […] We do reserve the right however, to modify or amend our opinion of the witness provided information, should the results of the official investigation provide significant evidence supporting other ignition scenarios related to human activities of any kind.

[32] As the trial judge pointed out, Matte’s premise as to the absence of human activity in the basement on the morning of the fire was ultimately false. The appellant admitted in his trial testimony that he did in fact go into the basement on the morning of the fire. Again, he was the only one in the home that morning. The testimony adduced at trial therefore significantly calls into question the accuracy of Matte’s conclusions, which undermines the appellant’s theory that Matte’s report or his testimony “probably” would have changed the result.

[33] In any event, there was no necessary inconsistency between the Matte Report and Williams’ conclusion that the fire was caused by the introduction of butane vapour into the basement environment. Matte focused on the ignition that triggered the fire without identifying a fuel source. In contrast, Williams identified a fuel source – butane vapour – but did not find it necessary to identify an ignition source since, regardless of what ignited the butane, it was the introduction of butane vapour that was the cause of the combustion. Thus, Matte’s conclusion that the fire had been ignited by the faulty extension cord connection may have been correct, and its accuracy would not call into question the reliability of Williams’ conclusion that the fire was caused by the introduction of butane vapour. In short, the two reports can be read in a consistent manner: one focuses on the ignition source and the other focuses on the fuel.

[34] As noted above, the appellant must do more than simply show that the result could possibly have been different if counsel had performed competently. Prejudice is established only where there is a reasonable probability that the outcome would have been different but for trial counsel’s alleged incompetent representation.

[35] Given the flaws and caveats in the Matte Report, particularly its failure to grapple with the fuel source of the fire, the trial judge found – correctly, in my view – that even if Matte had testified, it would not have affected the result. At the end of the day, the trial judge considered the opinions in the Matte Report despite its inadmissibility. The appellant has not introduced any doubt into the facts which supported the verdict’s foundation, since the defence theory has not changed since the trial and the trial judge considered that theory in its totality. Nor has the appellant identified any other expert whose evidence might have called into question the reliability of Williams’ conclusion that the introduction of butane vapour was the cause of the fire. Therefore, the appellant has not established a reasonable probability that trial counsel’s failure to seek legal aid funding for an expert would have affected the result.

[36] Since the appellant has failed to show that trial counsel’s representation resulted in a miscarriage of justice on either branch of the test, it is unnecessary to consider the performance component (i.e., whether trial counsel’s acts or omissions amounted to incompetence, measured against a reasonableness standard).

V. DISPOSITION

[37] For the above reasons, I would dismiss the appeal. Although the appellant originally also brought an appeal against sentence, he made no written or oral submissions regarding his sentence, which he advised has been fully served. His sentence appeal is therefore dismissed as abandoned.

Released: May 7, 2026 “J.M.F.”

“P.J. Monahan J.A.”

“I agree. Fairburn A.C.J.O.”

“I agree. P.J. Osborne J.A.”